A desktop coding agent with free access to frontier models.
TypeScript
{"subconscious":{"id":"subconscious","env":["SUBCONSCIOUS_API_KEY"],"npm":"@ai-sdk/anthropic","api":"https://api.subconscious.dev/v1","name":"Subconscious","doc":"https://docs.subconscious.dev","models":{"subconscious/glm-5.2":{"id":"subconscious/glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"subconscious/tim-qwen3.6-27b":{"id":"subconscious/tim-qwen3.6-27b","name":"TIM-Qwen3.6 27B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-05-11","last_updated":"2026-05-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"input":8192,"output":5000},"cost":{"input":0.3,"output":3,"cache_read":0.15}}}},"tokengo":{"id":"tokengo","env":["TOKENGO_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.tokengo.com/v1","name":"TokenGo","doc":"https://www.tokengo.com/docs","models":{"qwen/qwen3.5-397b-a17b":{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen3.5 397B-A17B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.4,"output":2.65,"cache_read":0.2}},"minimax/minimax-m2.5":{"id":"minimax/minimax-m2.5","name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03}},"deepseek/deepseek-v4-flash":{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.098,"output":0.196,"cache_read":0.028}},"deepseek/deepseek-v3.2":{"id":"deepseek/deepseek-v3.2","name":"DeepSeek V3.2","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":64000},"cost":{"input":0.2174,"output":0.326,"cache_read":0.06}},"deepseek/deepseek-v4-pro":{"id":"deepseek/deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.435,"output":0.87}},"deepseek/deepseek-v3.1":{"id":"deepseek/deepseek-v3.1","name":"DeepSeek-V3.1","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2025-08-21","last_updated":"2025-08-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":8192},"cost":{"input":0.19,"output":0.71,"cache_read":0.06}},"moonshotai/kimi-k2.6":{"id":"moonshotai/kimi-k2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.16}},"moonshotai/kimi-k3":{"id":"moonshotai/kimi-k3","name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3}},"z-ai/glm-5.2":{"id":"z-ai/glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"z-ai/glm-5.3-flash":{"id":"z-ai/glm-5.3-flash","name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0.075,"output":0.025,"cache_read":0.015}},"z-ai/glm-5":{"id":"z-ai/glm-5","name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.89,"output":3.2647,"cache_read":0.2226}},"z-ai/glm-5.1":{"id":"z-ai/glm-5.1","name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"z-ai/glm-5.3":{"id":"z-ai/glm-5.3","name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}}}},"modelis":{"id":"modelis","env":["MODELIS_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://modelishub.com/v1","name":"Modelis","doc":"https://modelishub.com/pricing","models":{"claude-sonnet-4-6":{"id":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15}},"deepseek-v4-flash":{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.0983,"output":0.1966}},"claude-fable-5":{"id":"claude-fable-5","name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50}},"claude-opus-4-8":{"id":"claude-opus-4-8","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25}},"deepseek-v4-pro":{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.435,"output":0.87}},"gemini-2.5-pro":{"id":"gemini-2.5-pro","name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10}},"gemini-2.5-flash":{"id":"gemini-2.5-flash","name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","max"]},{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5}},"qwen/qwen3.7-max":{"id":"qwen/qwen3.7-max","name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":3,"output":9}},"qwen/qwen3.7-plus":{"id":"qwen/qwen3.7-plus","name":"Qwen3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":0.768,"output":3.072}}}},"bothub":{"id":"bothub","env":["BOTHUB_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://openai.bothub.ru/v1","name":"Bothub","doc":"https://bothub.ru/models","models":{"nemotron-3-ultra-550b-a55b:free":{"id":"nemotron-3-ultra-550b-a55b:free","name":"Nemotron 3 Ultra (free)","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","medium","high"]}],"tool_call":true,"interleaved":true,"temperature":true,"release_date":"2026-06-04","last_updated":"2026-06-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":128000},"cost":{"input":0,"output":0}},"gemma-4-31b-it:free":{"id":"gemma-4-31b-it:free","name":"Gemma 4 31B IT (free)","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0}}}},"greenpt":{"id":"greenpt","env":["GREENPT_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.greenpt.ai/v1","name":"GreenPT","doc":"https://docs.greenpt.ai","models":{"deepseek-v4-flash-0731":{"id":"deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.1596,"output":0.399,"cache_read":0.0456}},"glm-5.2-caveman-ultra":{"id":"glm-5.2-caveman-ultra","name":"GLM-5.2 Caveman Ultra","description":"glm-5.2 carrying a built-in ruleset that compresses prose, keeping code and technical detail verbatim. The most aggressive tier, close to answer-only. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"glm-5.2-ponytail-ultra":{"id":"glm-5.2-ponytail-ultra","name":"GLM-5.2 Ponytail Ultra","description":"glm-5.2 carrying a built-in ruleset that compresses generated code, preferring platform features over custom code. The most aggressive tier, close to answer-only. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"glm-5.2-honey-ultra":{"id":"glm-5.2-honey-ultra","name":"GLM-5.2 Honey Ultra","description":"glm-5.2 carrying a built-in ruleset that compresses both generated code and prose. The most aggressive tier, close to answer-only. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"kimi-k2.6":{"id":"kimi-k2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.7524,"output":4.275,"cache_read":0.2508}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"minimax-m2.5":{"id":"minimax-m2.5","name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.1938,"output":1.129,"cache_read":0.0627}},"kimi-k2.7-code":{"id":"kimi-k2.7-code","name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.9006,"output":4.389,"cache_read":0.1881}},"green-l":{"id":"green-l","name":"Green L","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-03","release_date":"2025-06-20","last_updated":"2025-06-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32768},"cost":{"input":0.285,"output":0.912}},"devstral-2-123b-instruct-2512":{"id":"devstral-2-123b-instruct-2512","name":"Devstral 2","description":"Mistral's coding-agent model for repository work, terminal tasks, and software fixes","family":"devstral","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-12","release_date":"2025-12-09","last_updated":"2025-12-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":16384},"cost":{"input":0.57,"output":2.736}},"glm-5.2-honey-lite":{"id":"glm-5.2-honey-lite","name":"GLM-5.2 Honey Lite","description":"glm-5.2 carrying a built-in ruleset that compresses both generated code and prose. The gentlest tier: it cuts filler only and keeps the explanation intact. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"qwen3-coder-30b-a3b-instruct":{"id":"qwen3-coder-30b-a3b-instruct","name":"Qwen3-Coder 30B-A3B Instruct","description":"Smaller Qwen coder for efficient local agents and repo-level fixes","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32768},"cost":{"input":0.285,"output":1.083}},"green-l-raw":{"id":"green-l-raw","name":"Green L Raw","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-03","release_date":"2025-06-20","last_updated":"2025-06-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32768},"cost":{"input":0.285,"output":0.912}},"qwen3.5-397b-a17b":{"id":"qwen3.5-397b-a17b","name":"Qwen3.5 397B-A17B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":16384},"cost":{"input":0.798,"output":4.959}},"glm-5.2-caveman":{"id":"glm-5.2-caveman","name":"GLM-5.2 Caveman","description":"glm-5.2 carrying a built-in ruleset that compresses prose, keeping code and technical detail verbatim. The middle tier, and the ruleset as its authors wrote it. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"kimi-k3":{"id":"kimi-k3","name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":3.762,"output":18.81,"cache_read":0.9405}},"gemma-3-27b-it":{"id":"gemma-3-27b-it","name":"Gemma 3 27B","description":"Google Gemma 3 multimodal model for chat, reasoning, and image understanding","family":"gemma","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":40000,"output":8192},"cost":{"input":0.342,"output":0.684}},"qwen3.6-35b-a3b":{"id":"qwen3.6-35b-a3b","name":"Qwen3.6 35B-A3B","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.342,"output":2.052}},"green-s":{"id":"green-s","name":"Green S","description":"GreenPT speech-to-text model for pre-recorded and live transcription","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-01","last_updated":"2025-01","modalities":{"input":["audio"],"output":["text"]},"open_weights":false,"limit":{"context":0,"output":8192},"cost":{"input":0.00437,"output":0}},"glm-5.2-caveman-lite":{"id":"glm-5.2-caveman-lite","name":"GLM-5.2 Caveman Lite","description":"glm-5.2 carrying a built-in ruleset that compresses prose, keeping code and technical detail verbatim. The gentlest tier: it cuts filler only and keeps the explanation intact. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"mistral-medium-3.5-128b":{"id":"mistral-medium-3.5-128b","name":"Mistral Medium 3.5","description":"Balanced Mistral model for enterprise assistants, multilingual work, and tools","family":"mistral-medium","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-29","last_updated":"2026-04-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":16384},"cost":{"input":2.052,"output":10.26}},"glm-5.2-ponytail":{"id":"glm-5.2-ponytail","name":"GLM-5.2 Ponytail","description":"glm-5.2 carrying a built-in ruleset that compresses generated code, preferring platform features over custom code. The middle tier, and the ruleset as its authors wrote it. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"gemma4":{"id":"gemma4","name":"gemma4","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.57,"output":1.71}},"mistral-small-3.2-24b-instruct-2506":{"id":"mistral-small-3.2-24b-instruct-2506","name":"Mistral Small 3.2","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-03","release_date":"2025-06-20","last_updated":"2025-06-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32768},"cost":{"input":0.228,"output":0.456}},"glm-5.2-honey":{"id":"glm-5.2-honey","name":"GLM-5.2 Honey","description":"glm-5.2 carrying a built-in ruleset that compresses both generated code and prose. The middle tier, and the ruleset as its authors wrote it. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"glm-5.1":{"id":"glm-5.1","name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":131072},"status":"deprecated","cost":{"input":1.756,"output":5.518}},"glm-5.2-ponytail-lite":{"id":"glm-5.2-ponytail-lite","name":"GLM-5.2 Ponytail Lite","description":"glm-5.2 carrying a built-in ruleset that compresses generated code, preferring platform features over custom code. The gentlest tier: it cuts filler only and keeps the explanation intact. Same upstream model and price per token as glm-5.2, with fewer output tokens.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.254,"output":5.016,"cache_read":0.3135}},"qwen3-235b-a22b-instruct-2507":{"id":"qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507","description":"Qwen3 235B MoE instruct model for long-context multilingual chat and reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-21","last_updated":"2025-07-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":16384},"cost":{"input":1.026,"output":3.078}},"green-r-raw":{"id":"green-r-raw","name":"Green R Raw","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.399,"output":1.083}},"gpt-oss-120b":{"id":"gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.228,"output":0.798}},"holo2-30b-a3b":{"id":"holo2-30b-a3b","name":"Holo2 30B A3B","description":"H Company Holo2 vision model for GUI navigation and computer-use agents","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-11","last_updated":"2025-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":22016,"output":16384},"cost":{"input":0.399,"output":0.969}},"voxtral-small-24b-2507":{"id":"voxtral-small-24b-2507","name":"Voxtral Small 24B","description":"Mistral Voxtral audio-understanding model for speech and transcription tasks","family":"mistral","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-07","release_date":"2025-07-15","last_updated":"2025-07-15","modalities":{"input":["text","audio"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"output":16384},"cost":{"input":0.228,"output":0.513}},"green-s-pro":{"id":"green-s-pro","name":"Green S Pro","description":"GreenPT advanced speech-to-text model with multilingual transcription support","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-02","last_updated":"2025-02","modalities":{"input":["audio"],"output":["text"]},"open_weights":false,"limit":{"context":0,"output":8192},"cost":{"input":0.00437,"output":0}},"kimi-k2.6-fast":{"id":"kimi-k2.6-fast","name":"Kimi K2.6 Fast","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"status":"deprecated","cost":{"input":1.655,"output":8.778}},"pixtral-12b-2409":{"id":"pixtral-12b-2409","name":"Pixtral 12B","description":"Mistral vision-language model for image understanding and multimodal chat","family":"pixtral","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-09","release_date":"2024-09-01","last_updated":"2024-09-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0.285,"output":0.285}},"green-r":{"id":"green-r","name":"Green R","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.399,"output":1.083}},"llama-3.3-70b-instruct":{"id":"llama-3.3-70b-instruct","name":"Llama-3.3-70B-Instruct","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":100000,"output":16384},"cost":{"input":1.254,"output":1.254}}}},"qiniu-ai":{"id":"qiniu-ai","env":["QINIU_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.qnaigc.com/v1","name":"Qiniu","doc":"https://developer.qiniu.com/aitokenapi","models":{"glm-4.5-air":{"id":"glm-4.5-air","name":"GLM 4.5 Air","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":4096}},"kling-v2-6":{"id":"kling-v2-6","name":"Kling-V2 6","description":"Video model for prompt-guided generation, editing, and motion workflows","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-01-13","last_updated":"2026-01-13","modalities":{"input":["text","image","video"],"output":["video"]},"open_weights":false,"limit":{"context":99999999,"output":99999999}},"qwen3-next-80b-a3b-thinking":{"id":"qwen3-next-80b-a3b-thinking","name":"Qwen3 Next 80B A3B Thinking","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-12","last_updated":"2025-09-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":32768}},"deepseek-v3":{"id":"deepseek-v3","name":"DeepSeek-V3","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-08-13","last_updated":"2025-08-13","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16000}},"qwen3-235b-a22b-thinking-2507":{"id":"qwen3-235b-a22b-thinking-2507","name":"Qwen3 235B A22B Thinking 2507","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-12","last_updated":"2025-08-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":4096}},"gemini-3.0-pro-image-preview":{"id":"gemini-3.0-pro-image-preview","name":"Gemini 3.0 Pro Image Preview","description":"Image model for prompt-driven generation, editing, and visual design workflows","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-11-20","last_updated":"2025-11-20","modalities":{"input":["text","image"],"output":["text","image"]},"open_weights":false,"limit":{"context":32768,"output":8192}},"gemini-2.5-flash-image":{"id":"gemini-2.5-flash-image","name":"Gemini 2.5 Flash Image","description":"Image model for prompt-driven generation, editing, and visual design workflows","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-10-22","last_updated":"2025-10-22","modalities":{"input":["text","image"],"output":["image"]},"open_weights":false,"limit":{"context":32768,"output":8192}},"qwen3-next-80b-a3b-instruct":{"id":"qwen3-next-80b-a3b-instruct","name":"Qwen3 Next 80B A3B Instruct","description":"Tool-capable chat model for instruction following and agentic application workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-12","last_updated":"2025-09-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":32768}},"gemini-2.0-flash":{"id":"gemini-2.0-flash","name":"Gemini 2.0 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":8192}},"claude-3.5-sonnet":{"id":"claude-3.5-sonnet","name":"Claude 3.5 Sonnet","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-09","last_updated":"2025-09-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8200}},"doubao-seed-2.0-mini":{"id":"doubao-seed-2.0-mini","name":"Doubao Seed 2.0 Mini","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000}},"qwen-turbo":{"id":"qwen-turbo","name":"Qwen-Turbo","description":"Efficient Qwen model for fast chat, extraction, and high-volume workloads","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":4096}},"gemini-2.5-flash-lite":{"id":"gemini-2.5-flash-lite","name":"Gemini 2.5 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":64000}},"gemini-3.0-pro-preview":{"id":"gemini-3.0-pro-preview","name":"Gemini 3.0 Pro Preview","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-11-19","last_updated":"2025-11-19","modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000}},"doubao-seed-1.6":{"id":"doubao-seed-1.6","name":"Doubao-Seed 1.6","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-15","last_updated":"2025-08-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000}},"qwen3-32b":{"id":"qwen3-32b","name":"Qwen3 32B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":40000,"output":4096}},"mimo-v2-flash":{"id":"mimo-v2-flash","name":"Mimo-V2-Flash","description":"MiMo flash model for fast multimodal assistance and agent workflows","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-12-01","release_date":"2025-12-16","last_updated":"2026-02-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":256000},"cost":{"input":0.1,"output":0.3,"cache_read":0.01}},"gemini-2.0-flash-lite":{"id":"gemini-2.0-flash-lite","name":"Gemini 2.0 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":8192}},"claude-4.5-opus":{"id":"claude-4.5-opus","name":"Claude 4.5 Opus","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-11-25","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":200000}},"gpt-oss-20b":{"id":"gpt-oss-20b","name":"gpt-oss-20b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-06","last_updated":"2025-08-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096}},"qwen2.5-vl-72b-instruct":{"id":"qwen2.5-vl-72b-instruct","name":"Qwen 2.5 VL 72B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":8192}},"claude-4.0-opus":{"id":"claude-4.0-opus","name":"Claude 4.0 Opus","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000}},"qwen3-max-preview":{"id":"qwen3-max-preview","name":"Qwen3 Max Preview","description":"Flagship model for demanding analysis, coding, and production agent workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-06","last_updated":"2025-09-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":64000}},"doubao-1.5-pro-32k":{"id":"doubao-1.5-pro-32k","name":"Doubao 1.5 Pro 32k","description":"Flagship model for demanding analysis, coding, and production agent workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":12000}},"doubao-seed-2.0-code":{"id":"doubao-seed-2.0-code","name":"Doubao Seed 2.0 Code","description":"Coding model for repository understanding, refactors, and agentic engineering tasks","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":128000}},"qwen3.5-397b-a17b":{"id":"qwen3.5-397b-a17b","name":"Qwen3.5 397B A17B","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-22","last_updated":"2026-02-22","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":64000}},"deepseek-r1":{"id":"deepseek-r1","name":"DeepSeek-R1","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"qwen2.5-vl-7b-instruct":{"id":"qwen2.5-vl-7b-instruct","name":"Qwen 2.5 VL 7B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":8192}},"claude-4.1-opus":{"id":"claude-4.1-opus","name":"Claude 4.1 Opus","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-06","last_updated":"2025-08-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000}},"qwen3-30b-a3b-thinking-2507":{"id":"qwen3-30b-a3b-thinking-2507","name":"Qwen3 30b A3b Thinking 2507","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-04","last_updated":"2026-02-04","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":126000,"output":32000}},"deepseek-r1-0528":{"id":"deepseek-r1-0528","name":"DeepSeek-R1-0528","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"qwen-vl-max-2025-01-25":{"id":"qwen-vl-max-2025-01-25","name":"Qwen VL-MAX-2025-01-25","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096}},"doubao-seed-2.0-pro":{"id":"doubao-seed-2.0-pro","name":"Doubao Seed 2.0 Pro","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":128000}},"qwen3-max":{"id":"qwen3-max","name":"Qwen3 Max","description":"Flagship model for demanding analysis, coding, and production agent workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-24","last_updated":"2025-09-24","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":65536}},"glm-4.5":{"id":"glm-4.5","name":"GLM 4.5","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":98304}},"claude-3.5-haiku":{"id":"claude-3.5-haiku","name":"Claude 3.5 Haiku","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-26","last_updated":"2025-08-26","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192}},"doubao-seed-1.6-thinking":{"id":"doubao-seed-1.6-thinking","name":"Doubao-Seed 1.6 Thinking","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-15","last_updated":"2025-08-15","modalities":{"input":["image","text","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000}},"doubao-1.5-vision-pro":{"id":"doubao-1.5-vision-pro","name":"Doubao 1.5 Vision Pro","description":"Flagship model for demanding analysis, coding, and production agent workflows","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16000}},"doubao-seed-1.6-flash":{"id":"doubao-seed-1.6-flash","name":"Doubao-Seed 1.6 Flash","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-15","last_updated":"2025-08-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000}},"MiniMax-M1":{"id":"MiniMax-M1","name":"MiniMax M1","description":"MiniMax model for chat, coding, office work, and agentic tasks","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":80000}},"qwen3-30b-a3b-instruct-2507":{"id":"qwen3-30b-a3b-instruct-2507","name":"Qwen3 30b A3b Instruct 2507","description":"Tool-capable chat model for instruction following and agentic application workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-04","last_updated":"2026-02-04","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"qwen3-30b-a3b":{"id":"qwen3-30b-a3b","name":"Qwen3 30B A3B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":40000,"output":4096}},"qwen3-vl-30b-a3b-thinking":{"id":"qwen3-vl-30b-a3b-thinking","name":"Qwen3-Vl 30b A3b Thinking","description":"Multimodal model for analyzing text, images, documents, and rich media","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-09","last_updated":"2026-02-09","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"kimi-k2":{"id":"kimi-k2","name":"Kimi K2","description":"Kimi model for long-context chat, coding, and agentic reasoning","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":128000}},"claude-3.7-sonnet":{"id":"claude-3.7-sonnet","name":"Claude 3.7 Sonnet","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":128000}},"qwen3-235b-a22b":{"id":"qwen3-235b-a22b","name":"Qwen 3 235B A22B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"qwen3-235b-a22b-instruct-2507":{"id":"qwen3-235b-a22b-instruct-2507","name":"Qwen3 235b A22B Instruct 2507","description":"Tool-capable chat model for instruction following and agentic application workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-12","last_updated":"2025-08-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":64000}},"deepseek-v3-0324":{"id":"deepseek-v3-0324","name":"DeepSeek-V3-0324","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16000}},"doubao-1.5-thinking-pro":{"id":"doubao-1.5-thinking-pro","name":"Doubao 1.5 Thinking Pro","description":"Flagship model for demanding analysis, coding, and production agent workflows","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16000}},"gemini-3.0-flash-preview":{"id":"gemini-3.0-flash-preview","name":"Gemini 3.0 Flash Preview","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-18","last_updated":"2025-12-18","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000}},"claude-4.5-sonnet":{"id":"claude-4.5-sonnet","name":"Claude 4.5 Sonnet","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-30","last_updated":"2025-09-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000}},"gpt-oss-120b":{"id":"gpt-oss-120b","name":"gpt-oss-120b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-06","last_updated":"2025-08-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096}},"qwen-max-2025-01-25":{"id":"qwen-max-2025-01-25","name":"Qwen2.5-Max-2025-01-25","description":"Flagship Qwen model for complex reasoning, coding, and agentic workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096}},"qwen3-coder-480b-a35b-instruct":{"id":"qwen3-coder-480b-a35b-instruct","name":"Qwen3 Coder 480B A35B Instruct","description":"Coding model for repository understanding, refactors, and agentic engineering tasks","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-14","last_updated":"2025-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":4096}},"gemini-2.5-pro":{"id":"gemini-2.5-pro","name":"Gemini 2.5 Pro","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536}},"doubao-seed-2.0-lite":{"id":"doubao-seed-2.0-lite","name":"Doubao Seed 2.0 Lite","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000}},"claude-4.5-haiku":{"id":"claude-4.5-haiku","name":"Claude 4.5 Haiku","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-10-16","last_updated":"2025-10-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000}},"gemini-2.5-flash":{"id":"gemini-2.5-flash","name":"Gemini 2.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":64000}},"claude-4.0-sonnet":{"id":"claude-4.0-sonnet","name":"Claude 4.0 Sonnet","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000}},"deepseek-v3.1":{"id":"deepseek-v3.1","name":"DeepSeek-V3.1","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-19","last_updated":"2025-08-19","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"stepfun-ai/gelab-zero-4b-preview":{"id":"stepfun-ai/gelab-zero-4b-preview","name":"Stepfun-Ai/Gelab Zero 4b Preview","description":"StepFun flash model for efficient multimodal reasoning, coding, and tool use","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"output":4096}},"meituan/longcat-flash-lite":{"id":"meituan/longcat-flash-lite","name":"Meituan/Longcat-Flash-Lite","description":"Efficient model for low-latency assistance, extraction, and routine automation","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-06","last_updated":"2026-02-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":320000}},"meituan/longcat-flash-chat":{"id":"meituan/longcat-flash-chat","name":"Meituan/Longcat-Flash-Chat","description":"Efficient model for low-latency assistance, extraction, and routine automation","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-11-05","last_updated":"2025-11-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":131072}},"stepfun/step-3.5-flash":{"id":"stepfun/step-3.5-flash","name":"Stepfun/Step-3.5 Flash","description":"StepFun flash model for efficient multimodal reasoning, coding, and tool use","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-02-02","last_updated":"2026-02-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":64000,"output":4096}},"xiaomi/mimo-v2-flash":{"id":"xiaomi/mimo-v2-flash","name":"Xiaomi/Mimo-V2-Flash","description":"MiMo flash model for fast multimodal assistance and agent workflows","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-12-01","release_date":"2025-12-16","last_updated":"2026-02-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":256000},"cost":{"input":0.1,"output":0.3,"cache_read":0.01}},"minimax/minimax-m2.1":{"id":"minimax/minimax-m2.1","name":"Minimax/Minimax-M2.1","description":"MiniMax model for chat, coding, office work, and agentic tasks","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":204800,"output":128000}},"minimax/minimax-m2":{"id":"minimax/minimax-m2","name":"Minimax/Minimax-M2","description":"MiniMax model for chat, coding, office work, and agentic tasks","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-10-28","last_updated":"2025-10-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":128000}},"minimax/minimax-m2.5":{"id":"minimax/minimax-m2.5","name":"Minimax/Minimax-M2.5","description":"MiniMax model for chat, coding, office work, and agentic tasks","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":204800,"output":128000}},"minimax/minimax-m2.5-highspeed":{"id":"minimax/minimax-m2.5-highspeed","name":"Minimax/Minimax-M2.5 Highspeed","description":"High-speed MiniMax model for low-latency coding and agent workflows","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":204800,"output":128000}},"deepseek/deepseek-math-v2":{"id":"deepseek/deepseek-math-v2","name":"Deepseek/Deepseek-Math-V2","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-12-04","last_updated":"2025-12-04","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":160000,"output":160000}},"deepseek/deepseek-v3.2-exp-thinking":{"id":"deepseek/deepseek-v3.2-exp-thinking","name":"DeepSeek/DeepSeek-V3.2-Exp-Thinking","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"deepseek/deepseek-v3.1-terminus-thinking":{"id":"deepseek/deepseek-v3.1-terminus-thinking","name":"DeepSeek/DeepSeek-V3.1-Terminus-Thinking","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-09-22","last_updated":"2025-09-22","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"deepseek/deepseek-v3.2-exp":{"id":"deepseek/deepseek-v3.2-exp","name":"DeepSeek/DeepSeek-V3.2-Exp","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"deepseek/deepseek-v3.1-terminus":{"id":"deepseek/deepseek-v3.1-terminus","name":"DeepSeek/DeepSeek-V3.1-Terminus","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-22","last_updated":"2025-09-22","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"deepseek/deepseek-v3.2-251201":{"id":"deepseek/deepseek-v3.2-251201","name":"Deepseek/DeepSeek-V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":32000}},"x-ai/grok-code-fast-1":{"id":"x-ai/grok-code-fast-1","name":"x-AI/Grok-Code-Fast 1","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-02","last_updated":"2025-09-02","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":10000}},"x-ai/grok-4-fast-reasoning":{"id":"x-ai/grok-4-fast-reasoning","name":"X-Ai/Grok-4-Fast-Reasoning","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-18","last_updated":"2025-12-18","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":2000000}},"x-ai/grok-4.1-fast-non-reasoning":{"id":"x-ai/grok-4.1-fast-non-reasoning","name":"X-Ai/Grok 4.1 Fast Non Reasoning","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-19","last_updated":"2025-12-19","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":2000000}},"x-ai/grok-4.1-fast-reasoning":{"id":"x-ai/grok-4.1-fast-reasoning","name":"X-Ai/Grok 4.1 Fast Reasoning","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-19","last_updated":"2025-12-19","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":20000000,"output":2000000}},"x-ai/grok-4-fast":{"id":"x-ai/grok-4-fast","name":"x-AI/Grok-4-Fast","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-20","last_updated":"2025-09-20","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":2000000}},"x-ai/grok-4-fast-non-reasoning":{"id":"x-ai/grok-4-fast-non-reasoning","name":"X-Ai/Grok-4-Fast-Non-Reasoning","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-18","last_updated":"2025-12-18","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":2000000}},"x-ai/grok-4.1-fast":{"id":"x-ai/grok-4.1-fast","name":"x-AI/Grok-4.1-Fast","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-11-20","last_updated":"2025-11-20","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":2000000}},"openai/gpt-5.2":{"id":"openai/gpt-5.2","name":"OpenAI/GPT-5.2","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000}},"openai/gpt-5":{"id":"openai/gpt-5","name":"OpenAI/GPT-5","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-19","last_updated":"2025-09-19","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000}},"moonshotai/kimi-k2-0905":{"id":"moonshotai/kimi-k2-0905","name":"Kimi K2 0905","description":"Kimi model for long-context chat, coding, and agentic reasoning","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-08","last_updated":"2025-09-08","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":100000}},"moonshotai/kimi-k2-thinking":{"id":"moonshotai/kimi-k2-thinking","name":"Kimi K2 Thinking","description":"Kimi reasoning model for long-horizon research, planning, and tool use","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-11-07","last_updated":"2025-11-07","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":100000}},"moonshotai/kimi-k2.5":{"id":"moonshotai/kimi-k2.5","name":"Moonshotai/Kimi-K2.5","description":"Kimi multimodal agent model for visual understanding, coding, and planning","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-01-28","last_updated":"2026-01-28","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000}},"z-ai/glm-4.7":{"id":"z-ai/glm-4.7","name":"Z-Ai/GLM 4.7","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":200000}},"z-ai/glm-4.6":{"id":"z-ai/glm-4.6","name":"Z-AI/GLM 4.6","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-10-11","last_updated":"2025-10-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":200000}},"z-ai/autoglm-phone-9b":{"id":"z-ai/autoglm-phone-9b","name":"Z-Ai/Autoglm Phone 9b","description":"GLM vision model for visual reasoning, documents, and multimodal agents","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":12800,"output":4096}},"z-ai/glm-5":{"id":"z-ai/glm-5","name":"Z-Ai/GLM 5","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":128000}}}},"ambient":{"id":"ambient","env":["AMBIENT_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.ambient.xyz/v1","name":"Ambient","doc":"https://ambient.xyz","models":{"ambient/large":{"id":"ambient/large","name":"Ambient Large","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":202752},"cost":{"input":0.6,"output":2,"cache_read":0.15,"cache_write":0}},"stepfun/step-3.7-flash":{"id":"stepfun/step-3.7-flash","name":"Step 3.7 Flash","description":"StepFun flash model for efficient multimodal reasoning, coding, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03-01","release_date":"2026-05-29","last_updated":"2026-05-29","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.19,"output":1.14,"cache_read":0.03,"cache_write":0}},"xiaomi/mimo-v2.5":{"id":"xiaomi/mimo-v2.5","name":"MiMo-V2.5","description":"MiMo omni model for text, image, video, audio, and agents","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.4,"output":2,"cache_read":0.08,"cache_write":0}},"zai-org/GLM-5.2-FP8":{"id":"zai-org/GLM-5.2-FP8","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":202752},"cost":{"input":1.2,"output":4.2,"cache_read":0.26,"cache_write":0}},"zai-org/GLM-5.1-FP8":{"id":"zai-org/GLM-5.1-FP8","name":"GLM 5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0,"cache_write":0}},"deepseek/deepseek-v4-flash-0731":{"id":"deepseek/deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":0.08,"output":0.18,"cache_read":0.016,"cache_write":0}},"deepseek/deepseek-v4-flash":{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":0.14,"output":0.28,"cache_read":0.028,"cache_write":0}},"moonshotai/kimi-k2.6":{"id":"moonshotai/kimi-k2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.2,"cache_write":0}},"moonshotai/kimi-k2.7-code":{"id":"moonshotai/kimi-k2.7-code","name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.69,"output":3.49,"cache_read":0.14,"cache_write":0}},"z-ai/glm-5.2":{"id":"z-ai/glm-5.2","name":"GLM-5.2","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":202752},"cost":{"input":0.6,"output":2,"cache_read":0.15,"cache_write":0}}}},"agentrouter":{"id":"agentrouter","env":["AGENTROUTER_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://agentrouter.org/v1","name":"AgentRouter","doc":"https://agentrouter.org/docs/opencode.html","models":{"gpt-5.6-sol":{"id":"gpt-5.6-sol","name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000}},"claude-opus-5":{"id":"claude-opus-5","name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"provider":{"npm":"@ai-sdk/anthropic","api":"https://agentrouter.org/v1"}},"claude-opus-4-8":{"id":"claude-opus-4-8","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"provider":{"npm":"@ai-sdk/anthropic","api":"https://agentrouter.org/v1"}}}},"xiaomi-token-plan-cn":{"id":"xiaomi-token-plan-cn","env":["XIAOMI_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://token-plan-cn.xiaomimimo.com/v1","name":"Xiaomi Token Plan (China)","doc":"https://platform.xiaomimimo.com/#/docs","models":{"mimo-v2.5-tts":{"id":"mimo-v2.5-tts","name":"MiMo-V2.5-TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"mimo","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["audio"]},"open_weights":true,"limit":{"context":8192,"output":8192},"cost":{"input":0,"output":0}},"mimo-v2.5-tts-voiceclone":{"id":"mimo-v2.5-tts-voiceclone","name":"MiMo-V2.5-TTS-VoiceClone","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"mimo","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["audio"]},"open_weights":true,"limit":{"context":8192,"output":8192},"cost":{"input":0,"output":0}},"mimo-v2-tts":{"id":"mimo-v2-tts","name":"MiMo-V2-TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"mimo","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["audio"]},"open_weights":true,"limit":{"context":8192,"output":8192},"cost":{"input":0,"output":0}},"mimo-v2.5-tts-voicedesign":{"id":"mimo-v2.5-tts-voicedesign","name":"MiMo-V2.5-TTS-VoiceDesign","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"mimo","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["audio"]},"open_weights":true,"limit":{"context":8192,"output":8192},"cost":{"input":0,"output":0}},"mimo-v2-pro":{"id":"mimo-v2-pro","name":"MiMo-V2-Pro","description":"Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"status":"deprecated","cost":{"input":0,"output":0,"cache_read":0}},"mimo-v2.5":{"id":"mimo-v2.5","name":"MiMo-V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0}},"mimo-v2.5-pro":{"id":"mimo-v2.5-pro","name":"MiMo-V2.5-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0}}}},"nano-gpt":{"id":"nano-gpt","env":["NANO_GPT_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://nano-gpt.com/api/v1","name":"NanoGPT","doc":"https://docs.nano-gpt.com","models":{"glm-4.1v-thinking-flashx":{"id":"glm-4.1v-thinking-flashx","name":"GLM 4.1V Thinking FlashX","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"glm","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-09","last_updated":"2025-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":64000,"input":64000,"output":8192},"cost":{"input":0.3,"output":0.3,"cache_read":0.15}},"claude-opus-4-thinking:8192":{"id":"claude-opus-4-thinking:8192","name":"Claude 4 Opus Thinking (8K)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"qwen3.7-max":{"id":"qwen3.7-max","name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":2.5,"output":7.5,"cache_read":0.5,"cache_write":3.125}},"longcat-2.0":{"id":"longcat-2.0","name":"LongCat 2.0","description":"Meituan LongCat-2.0, a reasoning model with tool calling and a 1M-token context window","family":"longcat","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":262144},"cost":{"input":0.75,"output":3,"cache_read":0.015}},"gemma-4-31b-it-garnet":{"id":"gemma-4-31b-it-garnet","name":"Garnet","description":"Garnet is a multimodal Gemma 4 31B creative finetune for expressive dialogue, long-form storytelling, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"exa-answer":{"id":"exa-answer","name":"Exa (Answer)","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-23","last_updated":"2025-06-04","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":4096,"input":4096,"output":4096},"cost":{"input":2.5,"output":2.5}},"gemini-2.0-pro-exp-02-05":{"id":"gemini-2.0-pro-exp-02-05","name":"Gemini 2.0 Pro 0205","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-02-05","last_updated":"2025-02-05","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":2097152,"input":2097152,"output":8192},"cost":{"input":1.989,"output":7.956,"cache_read":0.49725}},"Meta-Llama-3-1-8B-Instruct-FP8":{"id":"Meta-Llama-3-1-8B-Instruct-FP8","name":"Llama 3.1 8B (decentralized)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.02,"output":0.03,"cache_read":0.01}},"MiniMax-M2":{"id":"MiniMax-M2","name":"MiniMax M2","description":"Efficient open MiniMax model built for coding agents and tool-heavy workflows","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-10-27","last_updated":"2025-10-27","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":131072},"cost":{"input":0.17,"output":1.53,"cache_read":0.085}},"ernie-5.0-thinking-preview":{"id":"ernie-5.0-thinking-preview","name":"Ernie 5.0 Thinking Preview","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"ernie","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2025-11-18","last_updated":"2025-11-18","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":1,"output":3.5,"cache_read":0.5}},"mercury-coder-small":{"id":"mercury-coder-small","name":"Mercury Coder Small","description":"Model by Inception AI. A diffusion large language model that runs incredibly quickly (500+ tokens/second) while matching Claude 3.5 Haiku and GPT-4o-mini. 1st in speed on Copilot arena, and matching 2nd in quality.","family":"mercury","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.25,"output":1,"cache_read":0.125}},"gemma-4-26b-a4b-it-luminous":{"id":"gemma-4-26b-a4b-it-luminous","name":"Luminous Mirror","description":"Luminous Mirror is a Gemma 4 26B A4B multimodal mixture-of-experts fine-tune for creative writing, expressive dialogue, and roleplay.","family":"gemma","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"claude-sonnet-4-thinking:32768":{"id":"claude-sonnet-4-thinking:32768","name":"Claude 4 Sonnet Thinking (32K)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"gemma-4-26b-a4b-it-shadowsiren":{"id":"gemma-4-26b-a4b-it-shadowsiren","name":"Shadow Siren","description":"Shadow Siren is a Gemma 4 26B A4B multimodal mixture-of-experts fine-tune for creative writing, expressive dialogue, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"auto-model-premium":{"id":"auto-model-premium","name":"Auto model (Premium)","description":"Automatic model router for matching prompts to suitable backends and budgets","family":"auto","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-04-16","last_updated":"2024-06-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":1000000},"cost":{"input":9.996,"output":19.992,"cache_read":4.998}},"mistral-code-latest":{"id":"mistral-code-latest","name":"Mistral Code Latest","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"mistral","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":0.3,"output":0.9,"cache_read":0.15}},"qwen3.5-122b-a10b:thinking":{"id":"qwen3.5-122b-a10b:thinking","name":"Qwen3.5 122B A10B Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.437,"output":3.496,"cache_read":0.103788}},"doubao-seed-1-6-250615":{"id":"doubao-seed-1-6-250615","name":"Doubao Seed 1.6","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-06-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":16384},"cost":{"input":0.204,"output":0.51,"cache_read":0.102}},"Gemma-4-26B-A4B-MeroMero":{"id":"Gemma-4-26B-A4B-MeroMero","name":"Gemma 4 26B A4B MeroMero","description":"Gemma 4 26B A4B MeroMero is an NVFP4 multimodal mixture-of-experts fine-tune for emotive dialogue, relationship scenes, creative writing, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-08-26","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"qwen-max":{"id":"qwen-max","name":"Qwen 2.5 Max","description":"Flagship Qwen model for complex reasoning, coding, and agentic workflows","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2024-04-03","last_updated":"2025-01-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32000,"input":32000,"output":8192},"cost":{"input":1.5997,"output":6.392,"cache_read":0.79985}},"claw-low":{"id":"claw-low","name":"Claw Low","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-11","last_updated":"2026-05-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2,"cache_write":0.08333}},"doubao-seed-2-0-mini-260215":{"id":"doubao-seed-2-0-mini-260215","name":"Doubao Seed 2.0 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":32000},"cost":{"input":0.0493,"output":0.4845,"cache_read":0.02465}},"qwen3.5-27b":{"id":"qwen3.5-27b","name":"Qwen3.5 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":260096,"input":260096,"output":65536},"cost":{"input":0.27,"output":2.16,"cache_read":0.135}},"claude-opus-4-1-thinking:1024":{"id":"claude-opus-4-1-thinking:1024","name":"Claude 4.1 Opus Thinking (1K)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"gemma-4-31b-it-gemsicle":{"id":"gemma-4-31b-it-gemsicle","name":"Gemsicle","description":"Gemsicle is a multimodal Gemma 4 31B creative finetune for expressive dialogue, long-form storytelling, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"qwen3.8-27b":{"id":"qwen3.8-27b","name":"Qwen3.8 27B","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.15,"output":0.7,"cache_read":0.04}},"qwen3.5-35b-a3b":{"id":"qwen3.5-35b-a3b","name":"Qwen3.5 35B A3B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":260096,"input":260096,"output":65536},"cost":{"input":0.225,"output":1.8,"cache_read":0.1125}},"claude-haiku-4-5-20251001-thinking":{"id":"claude-haiku-4-5-20251001-thinking","name":"Claude Haiku 4.5 Thinking","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1}},"gemini-2.5-flash-preview-09-2025-thinking":{"id":"gemini-2.5-flash-preview-09-2025-thinking","name":"Gemini 2.5 Flash Preview (09/2025) – Thinking","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text","image","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"gemma-4-26b-a4b-it-opusdistill":{"id":"gemma-4-26b-a4b-it-opusdistill","name":"Opus Distill","description":"Opus Distill is a Gemma 4 26B A4B multimodal mixture-of-experts fine-tune for creative writing, expressive dialogue, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"glm-4-plus-0111":{"id":"glm-4-plus-0111","name":"GLM 4 Plus 0111","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"glm","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-02-19","last_updated":"2025-02-19","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":4096},"cost":{"input":9.996,"output":9.996,"cache_read":4.998}},"gemini-2.0-pro-reasoner":{"id":"gemini-2.0-pro-reasoner","name":"Gemini 2.0 Pro Reasoner","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-02-05","modalities":{"input":["text","audio"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":1.292,"output":4.998,"cache_read":0.323}},"Qwen3.5-27B-Queen-Derestricted":{"id":"Qwen3.5-27B-Queen-Derestricted","name":"Qwen3.5 27B Queen Derestricted","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2026-04-30","last_updated":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.306,"output":0.306,"cache_read":0.153}},"qwen-3.6-plus":{"id":"qwen-3.6-plus","name":"Qwen 3.6 Plus","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"qwen3.6","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":65536},"cost":{"input":0.325,"output":1.95,"cache_read":0.0325,"cache_write":0.40625}},"Gemma-4-31B-Cognitive-Unshackled":{"id":"Gemma-4-31B-Cognitive-Unshackled","name":"Gemma 4 31B Cognitive Unshackled","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2026-05-01","last_updated":"2026-05-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.306,"output":0.306,"cache_read":0.153}},"gemini-2.5-pro-preview-06-05":{"id":"gemini-2.5-pro-preview-06-05","name":"Gemini 2.5 Pro Preview 0605","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-06-05","last_updated":"2025-06-05","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2.5,"output":10,"cache_read":0.25}},"claude-sonnet-4-thinking:8192":{"id":"claude-sonnet-4-thinking:8192","name":"Claude 4 Sonnet Thinking (8K)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"qwen-turbo":{"id":"qwen-turbo","name":"Qwen Turbo","description":"Efficient Qwen model for fast chat, extraction, and high-volume workloads","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2024-11-01","last_updated":"2025-04-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":8192},"cost":{"input":0.04998,"output":0.2006,"cache_read":0.02499}},"qwen3.7-flash:thinking":{"id":"qwen3.7-flash:thinking","name":"Qwen3.7 Flash Thinking","description":"Lightweight multimodal Qwen model for high-throughput text, image, and video tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":983616,"input":983616,"output":65536},"cost":{"input":0.03,"output":0.13,"cache_read":0.006,"cache_write":0.038}},"qwen3.5-omni-plus":{"id":"qwen3.5-omni-plus","name":"Qwen3.5 Omni Plus","description":"Omni-modal model for text, vision, audio, and multimodal agent tasks","family":"qwen3.5","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-03-30","last_updated":"2026-03-30","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":false,"limit":{"context":983616,"input":983616,"output":65536},"cost":{"input":0,"output":0}},"gemini-2.5-flash-lite":{"id":"gemini-2.5-flash-lite","name":"Gemini 2.5 Flash Lite","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01}},"asi1-mini":{"id":"asi1-mini","name":"ASI1 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-03-25","last_updated":"2025-03-25","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":1,"output":1,"cache_read":0.5}},"gemini-2.5-pro-preview-03-25":{"id":"gemini-2.5-pro-preview-03-25","name":"Gemini 2.5 Pro Preview 0325","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-03-25","last_updated":"2025-03-25","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2.5,"output":10,"cache_read":0.25}},"claude-sonnet-4-5-20250929-thinking":{"id":"claude-sonnet-4-5-20250929-thinking","name":"Claude Sonnet 4.5 Thinking","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"glm-4-air-0111":{"id":"glm-4-air-0111","name":"GLM 4 Air 0111","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"glm","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-01-11","last_updated":"2025-01-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":4096},"cost":{"input":0.1394,"output":0.1394,"cache_read":0.0697}},"claude-opus-4-1-20250805":{"id":"claude-opus-4-1-20250805","name":"Claude 4.1 Opus","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"qwen25-vl-72b-instruct":{"id":"qwen25-vl-72b-instruct","name":"Qwen25 VL 72b","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-05-10","last_updated":"2025-05-10","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":32000,"input":32000,"output":115200},"cost":{"input":0.69989,"output":0.69989,"cache_read":0.349945}},"qwen3.5-flash:thinking":{"id":"qwen3.5-flash:thinking","name":"Qwen3.5 Flash Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.05}},"command-a-reasoning-08-2025":{"id":"command-a-reasoning-08-2025","name":"Cohere Command A (08/2025)","description":"Cohere reasoning model for multilingual enterprise agents, tools, and complex workflows","family":"command-a","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-06-01","release_date":"2025-08-21","last_updated":"2025-08-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":8192},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"phi-4-multimodal-instruct":{"id":"phi-4-multimodal-instruct","name":"Phi 4 Multimodal","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"phi","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.07,"output":0.11,"cache_read":0.035}},"claude-opus-4-thinking:32768":{"id":"claude-opus-4-thinking:32768","name":"Claude 4 Opus Thinking (32K)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"qwen3.5-flash":{"id":"qwen3.5-flash","name":"Qwen3.5 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.05}},"mistral-code-agent-latest":{"id":"mistral-code-agent-latest","name":"Mistral Code Agent Latest","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"mistral","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.4,"output":2,"cache_read":0.2}},"Qwen3.5-27B-BlueStar-v3-Derestricted":{"id":"Qwen3.5-27B-BlueStar-v3-Derestricted","name":"Qwen3.5 27B BlueStar v3 Derestricted","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2026-04-30","last_updated":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.306,"output":0.306,"cache_read":0.153}},"deepseek-reasoner":{"id":"deepseek-reasoner","name":"DeepSeek Reasoner","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek-thinking","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-09","release_date":"2025-12-01","last_updated":"2026-02-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":64000,"input":64000,"output":65536},"cost":{"input":0.4,"output":1.7,"cache_read":0.2}},"gemma-4-e4b-it":{"id":"gemma-4-e4b-it","name":"Gemma 4 E4B Instruct","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":0.04,"output":0.2,"cache_read":0.02}},"GLM-4.6-Derestricted-v5":{"id":"GLM-4.6-Derestricted-v5","name":"GLM 4.6 Derestricted v5","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":8192},"cost":{"input":0.4,"output":1.5,"cache_read":0.2}},"azure-gpt-4o":{"id":"azure-gpt-4o","name":"Azure gpt-4o","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2024-05-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"doubao-seed-1-6-flash-250615":{"id":"doubao-seed-1-6-flash-250615","name":"Doubao Seed 1.6 Flash","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-06-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":16384},"cost":{"input":0.0374,"output":0.374,"cache_read":0.0187}},"gemini-2.5-flash-lite-preview-06-17":{"id":"gemini-2.5-flash-lite-preview-06-17","name":"Gemini 2.5 Flash Lite Preview","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.15,"output":0.6,"cache_read":0.015}},"claude-opus-4-thinking":{"id":"claude-opus-4-thinking","name":"Claude 4 Opus Thinking","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"brave-research":{"id":"brave-research","name":"Brave (Research)","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-02-10","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":16384,"input":16384,"output":16384},"cost":{"input":5,"output":5}},"doubao-1.5-pro-256k":{"id":"doubao-1.5-pro-256k","name":"Doubao 1.5 Pro 256k","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-03-12","last_updated":"2025-03-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":16384},"cost":{"input":0.799,"output":1.445,"cache_read":0.3995}},"qwen3-coder-30b-a3b-instruct":{"id":"qwen3-coder-30b-a3b-instruct","name":"Qwen3 Coder 30B A3B Instruct","description":"Smaller Qwen coder for efficient local agents and repo-level fixes","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.05}},"glm-z1-airx":{"id":"glm-z1-airx","name":"GLM Z1 AirX","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"glm","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2025-04-15","last_updated":"2025-04-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32000,"input":32000,"output":16384},"cost":{"input":0.7,"output":0.7,"cache_read":0.35}},"ernie-5.1:thinking":{"id":"ernie-5.1:thinking","name":"ERNIE 5.1 Thinking","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"ernie","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2026-05-10","last_updated":"2026-05-10","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":119000,"input":119000,"output":64000},"cost":{"input":0.75,"output":3,"cache_read":0.75}},"qwen3.8-max:thinking":{"id":"qwen3.8-max:thinking","name":"Qwen3.8 Max Thinking","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"input":991000,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25,"cache_write":2.5}},"universal-summarizer":{"id":"universal-summarizer","name":"Universal Summarizer","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-23","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":30,"output":30}},"venice-uncensored":{"id":"venice-uncensored","name":"Venice Uncensored","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"venice","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-10-01","last_updated":"2025-02-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.4,"output":1.8,"cache_read":0.4}},"gemini-2.5-flash-preview-09-2025":{"id":"gemini-2.5-flash-preview-09-2025","name":"Gemini 2.5 Flash Preview (09/2025)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text","image","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"claude-opus-4-20250514":{"id":"claude-opus-4-20250514","name":"Claude 4 Opus","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"doubao-1.5-pro-32k":{"id":"doubao-1.5-pro-32k","name":"Doubao 1.5 Pro 32k","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-11-20","last_updated":"2025-01-22","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32000,"input":32000,"output":8192},"cost":{"input":0.1343,"output":0.3349,"cache_read":0.06715}},"gemma-4-26b-a4b-it-moonlight":{"id":"gemma-4-26b-a4b-it-moonlight","name":"Moonlight Dusk","description":"Moonlight Dusk is a Gemma 4 26B A4B multimodal mixture-of-experts fine-tune for creative writing, expressive dialogue, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"deepseek-chat-cheaper":{"id":"deepseek-chat-cheaper","name":"DeepSeek V3/Chat Cheaper","description":"Chat-tuned GPT model for conversational assistance, writing, and tool workflows","family":"deepseek","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2025-04-15","last_updated":"2025-04-15","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.1,"output":0.425,"cache_read":0.05}},"gemma-4-26b-a4b-it-darksoul":{"id":"gemma-4-26b-a4b-it-darksoul","name":"Dark Soul","description":"Dark Soul is a Gemma 4 26B A4B multimodal mixture-of-experts fine-tune for creative writing, expressive dialogue, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"gemini-2.5-flash-lite-preview-09-2025":{"id":"gemini-2.5-flash-lite-preview-09-2025","name":"Gemini 2.5 Flash Lite Preview (09/2025)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text","image","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01}},"Gemma-4-31B-Queen":{"id":"Gemma-4-31B-Queen","name":"Gemma 4 31B Queen","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2026-05-01","last_updated":"2026-05-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.306,"output":0.306,"cache_read":0.153}},"deepseek-r1-sambanova":{"id":"deepseek-r1-sambanova","name":"DeepSeek R1 Fast","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"deepseek","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-02-20","last_updated":"2025-02-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":4096},"cost":{"input":4.998,"output":6.987,"cache_read":2.499}},"deepseek-r1":{"id":"deepseek-r1","name":"DeepSeek R1","description":"Classic open reasoning model for transparent math, coding, and deliberate problem solving","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-07","release_date":"2025-01-20","last_updated":"2025-05-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.4,"output":1.7,"cache_read":0.2}},"perplexity-academic-researcher":{"id":"perplexity-academic-researcher","name":"Perplexity Academic Researcher","description":"Sonar Reasoning Pro with Perplexity's academic search mode. Prioritizes scholarly and peer-reviewed sources from academic repositories and returns cited research synthesis.","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2026-07-10","last_updated":"2026-07-10","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":115200},"cost":{"input":2,"output":8,"cache_read":1}},"claude-sonnet-4-thinking:64000":{"id":"claude-sonnet-4-thinking:64000","name":"Claude 4 Sonnet Thinking (64K)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"qwen3.7-max:thinking":{"id":"qwen3.7-max:thinking","name":"Qwen3.7 Max Thinking","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":2.5,"output":7.5,"cache_read":0.5,"cache_write":3.125}},"gemma-4-26b-a4b-uncensored":{"id":"gemma-4-26b-a4b-uncensored","name":"Gemma 4 26B A4B Uncensored","description":"Gemma 4 26B A4B Uncensored is an FP8 open-weight multimodal mixture-of-experts model LoRA-tuned for fewer refusals across chat, coding, tool use, and long-context work.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"claude-opus-4-5-20251101:thinking":{"id":"claude-opus-4-5-20251101:thinking","name":"Claude 4.5 Opus Thinking","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2025-11-01","last_updated":"2025-11-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5}},"qwen3.5-27b:thinking":{"id":"qwen3.5-27b:thinking","name":"Qwen3.5 27B Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":260096,"input":260096,"output":65536},"cost":{"input":0.27,"output":2.16,"cache_read":0.135}},"qwen3.5-0.8b":{"id":"qwen3.5-0.8b","name":"Qwen3.5 0.8B","description":"Qwen3.5 0.8B is a lightweight open-weight multimodal model from Alibaba for fast reasoning, visual understanding, tool use, and JSON output.","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-08-16","last_updated":"2026-08-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.06,"output":0.12,"cache_read":0.03}},"doubao-seed-2-0-code-preview-260215":{"id":"doubao-seed-2-0-code-preview-260215","name":"Doubao Seed 2.0 Code Preview","description":"ByteDance Seed coding model for multimodal software engineering and long-running agents","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":128000},"cost":{"input":0.782,"output":3.893,"cache_read":0.391}},"ernie-5.1":{"id":"ernie-5.1","name":"ERNIE 5.1","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"ernie","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-05-10","last_updated":"2026-05-10","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":119000,"input":119000,"output":64000},"cost":{"input":0.75,"output":3,"cache_read":0.75}},"azure-gpt-4o-mini":{"id":"azure-gpt-4o-mini","name":"Azure gpt-4o-mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2024-07-18","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"gemma-4-26b-a4b-it-chimerax":{"id":"gemma-4-26b-a4b-it-chimerax","name":"Chimera X","description":"Chimera X is a Gemma 4 26B A4B multimodal mixture-of-experts fine-tune for creative writing, expressive dialogue, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"claude-sonnet-4-5-20250929":{"id":"claude-sonnet-4-5-20250929","name":"Claude Sonnet 4.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"qwen3.7-flash":{"id":"qwen3.7-flash","name":"Qwen3.7 Flash","description":"Lightweight multimodal Qwen model for high-throughput text, image, and video tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":65536},"cost":{"input":0.03,"output":0.13,"cache_read":0.006,"cache_write":0.038}},"sarvam-105b":{"id":"sarvam-105b","name":"Sarvam 105B","description":"Flagship Indian-language reasoning model for enterprise multilingual applications","family":"sarvam","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":4096},"cost":{"input":0.045,"output":0.177,"cache_read":0.028}},"deepseek-chat":{"id":"deepseek-chat","name":"DeepSeek V3/Deepseek Chat","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-09","release_date":"2025-12-01","last_updated":"2026-02-28","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.1,"output":0.425,"cache_read":0.05}},"holo3-35b-a3b":{"id":"holo3-35b-a3b","name":"Holo3-35B-A3B","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":8192},"cost":{"input":0.25,"output":1.8,"cache_read":0.125}},"hermes-high":{"id":"hermes-high","name":"Hermes High","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"hermes","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-11","last_updated":"2026-05-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"claw-high":{"id":"claw-high","name":"Claw High","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-11","last_updated":"2026-05-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"holo3-35b-a3b:thinking":{"id":"holo3-35b-a3b:thinking","name":"Holo3-35B-A3B Thinking","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":8192},"cost":{"input":0.25,"output":1.8,"cache_read":0.125}},"qwen3.5-omni-flash":{"id":"qwen3.5-omni-flash","name":"Qwen3.5 Omni Flash","description":"Omni-modal model for text, vision, audio, and multimodal agent tasks","family":"qwen3.5","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-03-30","last_updated":"2026-03-30","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":false,"limit":{"context":49152,"input":49152,"output":16384},"cost":{"input":0,"output":0}},"azure-gpt-4-turbo":{"id":"azure-gpt-4-turbo","name":"Azure gpt-4-turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":4096},"cost":{"input":10,"output":30}},"qwen3.5-35b-a3b:thinking":{"id":"qwen3.5-35b-a3b:thinking","name":"Qwen3.5 35B A3B Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":260096,"input":260096,"output":65536},"cost":{"input":0.225,"output":1.8,"cache_read":0.1125}},"claude-sonnet-4-thinking:1024":{"id":"claude-sonnet-4-thinking:1024","name":"Claude 4 Sonnet Thinking (1K)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"doubao-1.5-vision-pro-32k":{"id":"doubao-1.5-vision-pro-32k","name":"Doubao 1.5 Vision Pro 32k","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-11-20","last_updated":"2025-01-22","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":32000,"input":32000,"output":8192},"cost":{"input":0.459,"output":1.377,"cache_read":0.2295}},"doubao-seed-2-0-lite-260215":{"id":"doubao-seed-2-0-lite-260215","name":"Doubao Seed 2.0 Lite","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":32000},"cost":{"input":0.1462,"output":0.8738,"cache_read":0.0731}},"Gemma-4-26B-A4B-MeroMero:thinking":{"id":"Gemma-4-26B-A4B-MeroMero:thinking","name":"Gemma 4 26B A4B MeroMero Thinking","description":"Gemma 4 26B A4B MeroMero with thinking enabled for more deliberate emotive dialogue, relationship scenes, creative writing, and multimodal roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-08-26","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"gemini-2.5-flash-preview-04-17:thinking":{"id":"gemini-2.5-flash-preview-04-17:thinking","name":"Gemini 2.5 Flash Preview Thinking","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-04-17","last_updated":"2025-04-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.15,"output":3.5,"cache_read":0.015}},"claude-haiku-4-5-20251001":{"id":"claude-haiku-4-5-20251001","name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1}},"claw-medium":{"id":"claw-medium","name":"Claw Medium","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-11","last_updated":"2026-05-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"glm-4-long":{"id":"glm-4-long","name":"GLM-4 Long","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"glm","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-08-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":4096},"cost":{"input":0.2006,"output":0.2006,"cache_read":0.1003}},"Gemma-4-31B-GarnetV2":{"id":"Gemma-4-31B-GarnetV2","name":"Gemma 4 31B Garnet V2","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2026-05-01","last_updated":"2026-05-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.306,"output":0.306,"cache_read":0.153}},"gemini-2.5-flash-preview-04-17":{"id":"gemini-2.5-flash-preview-04-17","name":"Gemini 2.5 Flash Preview","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-04-17","last_updated":"2025-04-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.15,"output":0.6,"cache_read":0.015}},"Gemma-4-31B-Claude-4.6-Opus-Reasoning-Distilled":{"id":"Gemma-4-31B-Claude-4.6-Opus-Reasoning-Distilled","name":"Gemma 4 31B Claude 4.6 Opus Reasoning Distilled","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"claude","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2026-05-01","last_updated":"2026-05-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.306,"output":0.306,"cache_read":0.0306}},"qwen3.8-27b:thinking":{"id":"qwen3.8-27b:thinking","name":"Qwen3.8 27B Thinking","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.15,"output":0.7,"cache_read":0.04}},"qwen-plus":{"id":"qwen-plus","name":"Qwen Plus","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2024-01-25","last_updated":"2025-09-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":995904,"input":995904,"output":32768},"cost":{"input":0.3995,"output":1.2002,"cache_read":0.19975}},"fastgpt":{"id":"fastgpt","name":"Web Answer","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-23","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":7.5,"output":7.5}},"gemini-2.5-flash-nothinking":{"id":"gemini-2.5-flash-nothinking","name":"Gemini 2.5 Flash (No Thinking)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-06-05","last_updated":"2025-06-05","modalities":{"input":["text","image","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"yi-medium-200k":{"id":"yi-medium-200k","name":"Yi Medium 200k","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"yi","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-03-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":4096},"cost":{"input":2.499,"output":2.499,"cache_read":1.2495}},"qwen3.5-122b-a10b":{"id":"qwen3.5-122b-a10b","name":"Qwen3.5 122B A10B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.437,"output":3.496,"cache_read":0.103788}},"deepseek-reasoner-cheaper":{"id":"deepseek-reasoner-cheaper","name":"Deepseek R1 Cheaper","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"deepseek","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-01-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.4,"output":1.7,"cache_read":0.2}},"sonar":{"id":"sonar","name":"Perplexity Simple","description":"Fast web-grounded Sonar for current answers, citations, and lightweight retrieval","family":"sonar","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":127072,"input":127072,"output":114364},"cost":{"input":1,"output":1,"cache_read":0.5}},"gemini-2.5-flash-lite-preview-09-2025-thinking":{"id":"gemini-2.5-flash-lite-preview-09-2025-thinking","name":"Gemini 2.5 Flash Lite Preview (09/2025) – Thinking","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text","image","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01}},"claude-sonnet-4-thinking":{"id":"claude-sonnet-4-thinking","name":"Claude 4 Sonnet Thinking","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"auto-model-standard":{"id":"auto-model-standard","name":"Auto model (Standard)","description":"Automatic model router for matching prompts to suitable backends and budgets","family":"auto","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-04-16","last_updated":"2024-06-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":1000000},"cost":{"input":9.996,"output":19.992,"cache_read":4.998}},"gemini-3-pro-image-preview":{"id":"gemini-3-pro-image-preview","name":"Gemini 3 Pro Image","description":"Nano Banana Pro for higher-fidelity image generation and design-heavy edits","family":"gemini-pro","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2025-11-20","last_updated":"2025-11-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2}},"sonar-reasoning-pro":{"id":"sonar-reasoning-pro","name":"Perplexity Reasoning Pro","description":"Web-grounded Sonar for multi-step research questions that need cited reasoning","family":"sonar-reasoning","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":115200},"cost":{"input":2,"output":8,"cache_read":1}},"MiniMax-M1":{"id":"MiniMax-M1","name":"MiniMax M1","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"minimax","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-01-08","last_updated":"2025-06-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":131072},"cost":{"input":0.1394,"output":1.3328,"cache_read":0.0697}},"qwen3-30b-a3b-instruct-2507":{"id":"qwen3-30b-a3b-instruct-2507","name":"Qwen3 30B A3B Instruct 2507","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-02-20","last_updated":"2025-02-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":0.2,"output":0.5,"cache_read":0.1}},"Gemma-4-31B-DarkIdol":{"id":"Gemma-4-31B-DarkIdol","name":"Gemma 4 31B DarkIdol","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"release_date":"2026-05-01","last_updated":"2026-05-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.306,"output":0.306,"cache_read":0.153}},"qwen3.7-plus:thinking":{"id":"qwen3.7-plus:thinking","name":"Qwen3.7 Plus Thinking","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":983616,"input":983616,"output":65536},"cost":{"input":0.4,"output":1.6,"cache_read":0.08,"cache_write":0.5}},"auto-model":{"id":"auto-model","name":"Auto model","description":"Automatic model router for matching prompts to suitable backends and budgets","family":"auto","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-04-16","last_updated":"2024-06-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":1000000},"cost":{"input":0,"output":0}},"gemma-4-31b-it-darkidol":{"id":"gemma-4-31b-it-darkidol","name":"DarkIdol","description":"DarkIdol is a multimodal Gemma 4 31B creative finetune for expressive dialogue, long-form storytelling, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"claude-opus-4-1-thinking":{"id":"claude-opus-4-1-thinking","name":"Claude 4.1 Opus Thinking","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"glm-4.1v-thinking-flash":{"id":"glm-4.1v-thinking-flash","name":"GLM 4.1V Thinking Flash","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"glm","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-09","last_updated":"2025-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":64000,"input":64000,"output":8192},"cost":{"input":0.3,"output":0.3,"cache_read":0.15}},"qwen3.5-4b":{"id":"qwen3.5-4b","name":"Qwen3.5 4B","description":"Qwen3.5 4B is a compact open-weight multimodal model from Alibaba for reasoning, coding, visual understanding, tool use, and structured output.","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-08-16","last_updated":"2026-08-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.2,"cache_read":0.05}},"claude-opus-4-1-thinking:32768":{"id":"claude-opus-4-1-thinking:32768","name":"Claude 4.1 Opus Thinking (32K)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"mistral-small-31-24b-instruct":{"id":"mistral-small-31-24b-instruct","name":"Mistral Small 31 24b Instruct","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"mistral-small","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-04-15","last_updated":"2025-04-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":102400},"cost":{"input":0.1,"output":0.3,"cache_read":0.05}},"gemma-4-31b-it-fabled":{"id":"gemma-4-31b-it-fabled","name":"Fabled","description":"Fabled is a multimodal Gemma 4 31B creative finetune for expressive dialogue, long-form storytelling, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"qwen3-vl-235b-a22b-instruct-original":{"id":"qwen3-vl-235b-a22b-instruct-original","name":"Qwen3 VL 235B A22B Instruct Original","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.5,"output":1.2,"cache_read":0.25}},"azure-o3-mini":{"id":"azure-o3-mini","name":"Azure o3-mini","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-01-31","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":65536},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"qwen3.6-max-preview":{"id":"qwen3.6-max-preview","name":"Qwen3.6 Max Preview","description":"Flagship Qwen model for complex reasoning, coding, and agentic workflows","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-20","last_updated":"2026-04-20","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":245760,"input":245760,"output":65536},"cost":{"input":1.04,"output":6.24,"cache_read":0.52}},"gemini-2.5-flash-preview-05-20":{"id":"gemini-2.5-flash-preview-05-20","name":"Gemini 2.5 Flash 0520","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-05-20","last_updated":"2025-05-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1048000,"input":1048000,"output":65536},"cost":{"input":0.15,"output":0.6,"cache_read":0.015}},"mercury-2":{"id":"mercury-2","name":"Mercury 2","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"mercury","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":50000},"cost":{"input":0.25,"output":0.75,"cache_read":0.025}},"gemma-4-31b-it-gembrain":{"id":"gemma-4-31b-it-gembrain","name":"Gembrain","description":"Gembrain is a multimodal Gemma 4 31B creative finetune for expressive dialogue, long-form storytelling, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"claude-opus-4-thinking:1024":{"id":"claude-opus-4-thinking:1024","name":"Claude 4 Opus Thinking (1K)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"qwen3.8-max":{"id":"qwen3.8-max","name":"Qwen3.8 Max","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"input":991000,"output":65536},"cost":{"input":2,"output":6,"cache_read":0.25,"cache_write":2.5}},"longcat-2.0:thinking":{"id":"longcat-2.0:thinking","name":"LongCat 2.0 Thinking","description":"Meituan LongCat-2.0, a reasoning model with tool calling and a 1M-token context window","family":"longcat","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":262144},"cost":{"input":0.75,"output":3,"cache_read":0.015}},"qwen3-vl-235b-a22b-thinking":{"id":"qwen3-vl-235b-a22b-thinking","name":"Qwen3 VL 235B A22B Thinking","description":"Qwen vision-language thinking model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.5,"output":6,"cache_read":0.25}},"celeris-1":{"id":"celeris-1","name":"Celeris 1","description":"Celeris 1 is a diffusion language model built for ultra-low-latency classification, extraction, judging, query rewriting, and other short structured responses.","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-07-25","last_updated":"2026-07-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"input":8192,"output":8192},"cost":{"input":2,"output":6,"cache_read":1}},"qwen3.7-plus":{"id":"qwen3.7-plus","name":"Qwen3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":65536},"cost":{"input":0.4,"output":1.6,"cache_read":0.08,"cache_write":0.5}},"deepseek-v3-0324":{"id":"deepseek-v3-0324","name":"DeepSeek Chat 0324","description":"March 2025 checkpoint of DeepSeek-V3 with improved reasoning and coding","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-03-24","last_updated":"2025-03-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.2,"output":0.77,"cache_read":0.135}},"brave-pro":{"id":"brave-pro","name":"Brave (Pro)","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-02-10","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"input":8192,"output":8192},"cost":{"input":5,"output":5}},"gemma-4-31b-it-novelist":{"id":"gemma-4-31b-it-novelist","name":"Novelist","description":"Novelist is a multimodal Gemma 4 31B creative finetune for expressive dialogue, long-form storytelling, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"gemma-4-12b-it":{"id":"gemma-4-12b-it","name":"Gemma 4 12B Instruct","description":"Google's Gemma 4 12B Instruct is an open-weight multimodal model for text, image, audio, and video understanding, with tool calling and structured output support.","family":"gemma","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-08-01","last_updated":"2026-08-01","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.05,"output":0.25,"cache_read":0.025}},"sonar-pro":{"id":"sonar-pro","name":"Perplexity Pro","description":"Deeper Sonar search model with broader retrieval and stronger synthesis","family":"sonar-pro","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":3,"output":15,"cache_read":1.5}},"qwen3.5-2b":{"id":"qwen3.5-2b","name":"Qwen3.5 2B","description":"Qwen3.5 2B is a small open-weight multimodal model from Alibaba for efficient reasoning, coding, visual understanding, tool use, and JSON output.","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-08-16","last_updated":"2026-08-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.08,"output":0.16,"cache_read":0.04}},"doubao-seed-2-0-pro-260215":{"id":"doubao-seed-2-0-pro-260215","name":"Doubao Seed 2.0 Pro","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":128000},"cost":{"input":0.782,"output":3.876,"cache_read":0.391}},"Gemma-4-31B-MeroMero-v2:thinking":{"id":"Gemma-4-31B-MeroMero-v2:thinking","name":"Gemma 4 31B MeroMero v2 Thinking","description":"Gemma 4 31B MeroMero v2 with thinking enabled for more deliberate emotive dialogue, relationship scenes, creative writing, and multimodal roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-08-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"sonar-deep-research":{"id":"sonar-deep-research","name":"Perplexity Deep Research","description":"Sonar search model for autonomous research and citation-backed long-form reports","family":"sonar","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2025-02-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":115200},"cost":{"input":3.4,"output":13.6,"cache_read":1.7}},"kimi-k2-instruct-fast":{"id":"kimi-k2-instruct-fast","name":"Kimi K2 0711 Fast","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"kimi-k2","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-12-15","last_updated":"2025-07-15","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":0.4,"output":1.8,"cache_read":0.2}},"Gemma-4-31B-MeroMero-v2":{"id":"Gemma-4-31B-MeroMero-v2","name":"Gemma 4 31B MeroMero v2","description":"Gemma 4 31B MeroMero v2 is a LoRA finetune for emotive dialogue, relationship scenes, creative writing, and multimodal roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-08-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"auto-model-basic":{"id":"auto-model-basic","name":"Auto model (Basic)","description":"Automatic model router for matching prompts to suitable backends and budgets","family":"auto","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-04-16","last_updated":"2024-06-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":1000000},"cost":{"input":9.996,"output":19.992,"cache_read":4.998}},"gemini-2.5-pro":{"id":"gemini-2.5-pro","name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"cache_write":0.375}},"gemini-2.5-flash-preview-05-20:thinking":{"id":"gemini-2.5-flash-preview-05-20:thinking","name":"Gemini 2.5 Flash 0520 Thinking","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-05-20","last_updated":"2025-05-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1048000,"input":1048000,"output":65536},"cost":{"input":0.15,"output":3.5,"cache_read":0.015}},"gemini-2.5-pro-exp-03-25":{"id":"gemini-2.5-pro-exp-03-25","name":"Gemini 2.5 Pro Experimental 0325","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-03-25","last_updated":"2025-03-25","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2.5,"output":10,"cache_read":0.25}},"claude-sonnet-4-20250514":{"id":"claude-sonnet-4-20250514","name":"Claude 4 Sonnet","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"qwen-long":{"id":"qwen-long","name":"Qwen Long 10M","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-08-01","last_updated":"2025-01-25","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":10000000,"input":10000000,"output":8192},"cost":{"input":0.1003,"output":0.408,"cache_read":0.05015}},"phi-4-mini-instruct":{"id":"phi-4-mini-instruct","name":"Phi 4 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"phi","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.17,"output":0.68,"cache_read":0.085}},"hermes-low":{"id":"hermes-low","name":"Hermes Low","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"hermes","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-11","last_updated":"2026-05-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2,"cache_write":0.08333}},"azure-o1":{"id":"azure-o1","name":"Azure o1","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-17","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5}},"nano-gpt-help":{"id":"nano-gpt-help","name":"NanoGPT Help","description":"Text-only NanoGPT support assistant. Questions are processed by the Help inference provider; do not paste secrets or account credentials. Covers the website, models, API, pricing, memory, media generation, and support.","family":"gpt","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-06-06","last_updated":"2026-06-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":6000,"input":6000,"output":512},"cost":{"input":0,"output":0}},"gemini-2.5-flash":{"id":"gemini-2.5-flash","name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"qwen3-max-2026-01-23":{"id":"qwen3-max-2026-01-23","name":"Qwen3 Max 2026-01-23","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-01-26","last_updated":"2026-01-26","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":1.2002,"output":6.001,"cache_read":0.6001}},"gemini-2.5-pro-preview-05-06":{"id":"gemini-2.5-pro-preview-05-06","name":"Gemini 2.5 Pro Preview 0506","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2025-05-06","last_updated":"2025-05-06","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2.5,"output":10,"cache_read":0.25}},"brave":{"id":"brave","name":"Brave (Answers)","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-02-13","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"input":8192,"output":8192},"cost":{"input":5,"output":5}},"gemma-4-26b-a4b-uncensored:thinking":{"id":"gemma-4-26b-a4b-uncensored:thinking","name":"Gemma 4 26B A4B Uncensored Thinking","description":"Gemma 4 26B A4B Uncensored with thinking enabled for more deliberate coding, multimodal analysis, tool use, and long-context problem solving.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"pokee-isaac":{"id":"pokee-isaac","name":"Pokee-Isaac 28B","description":"Pokee-Isaac is a 28B agentic model with a roughly 10-million-token context window, function calling, and OpenAI-compatible structured output. Pokee bills in $0.01 increments, rounding each non-zero request up to the next cent.","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-08-04","last_updated":"2026-08-04","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":10000000,"input":10000000,"output":60000},"cost":{"input":0.15,"output":1,"cache_read":0.075}},"claude-opus-4-1-thinking:8192":{"id":"claude-opus-4-1-thinking:8192","name":"Claude 4.1 Opus Thinking (8K)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5}},"ernie-x1.1-preview":{"id":"ernie-x1.1-preview","name":"ERNIE X1.1","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"ernie","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-10","last_updated":"2025-09-10","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":64000,"input":64000,"output":8192},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"gemma-4-e2b-it":{"id":"gemma-4-e2b-it","name":"Gemma 4 E2B Instruct","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":0.02,"output":0.1,"cache_read":0.01}},"gemini-exp-1206":{"id":"gemini-exp-1206","name":"Gemini 2.0 Pro 1206","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gemini","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":2097152,"input":2097152,"output":8192},"cost":{"input":1.258,"output":4.998,"cache_read":0.629}},"gemma-4-31b-it-isometry":{"id":"gemma-4-31b-it-isometry","name":"Isometry","description":"Isometry is a multimodal Gemma 4 31B creative finetune for expressive dialogue, long-form storytelling, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"gemma-4-26b-a4b-it-musica":{"id":"gemma-4-26b-a4b-it-musica","name":"Musica","description":"Musica is a Gemma 4 26B A4B multimodal mixture-of-experts fine-tune for creative writing, expressive dialogue, and roleplay.","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"yi-large":{"id":"yi-large","name":"Yi Large","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"yi","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-05-13","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32000,"input":32000,"output":4096},"cost":{"input":3.196,"output":3.196,"cache_read":1.598}},"hermes-medium":{"id":"hermes-medium","name":"Hermes Medium","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"hermes","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-11","last_updated":"2026-05-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"claude-opus-4-5-20251101":{"id":"claude-opus-4-5-20251101","name":"Claude 4.5 Opus","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2025-11-01","last_updated":"2025-11-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5}},"deepclaude":{"id":"deepclaude","name":"DeepClaude","description":"Compact GPT model for low-latency assistance and high-volume workloads","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-03-10","last_updated":"2025-02-01","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":3,"output":15,"cache_read":0.3}},"qvq-max":{"id":"qvq-max","name":"Qwen: QvQ Max","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-03-28","last_updated":"2025-03-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":1.2,"output":4.8,"cache_read":0.6}},"LatitudeGames/Wayfarer-Large-70B-Llama-3.3":{"id":"LatitudeGames/Wayfarer-Large-70B-Llama-3.3","name":"Llama 3.3 70B Wayfarer","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-02-20","last_updated":"2025-02-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.7,"output":0.7,"cache_read":0.35}},"qwen/qwen3.5-397b-a17b-thinking":{"id":"qwen/qwen3.5-397b-a17b-thinking","name":"Qwen3.5 397B A17B Thinking","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":258048,"input":258048,"output":65536},"cost":{"input":0.6,"output":3.6,"cache_read":0.3}},"qwen/qwen3-coder-plus":{"id":"qwen/qwen3-coder-plus","name":"Qwen3 Coder Plus","description":"Hosted Qwen coder for software agents, repo edits, and long-context code","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-23","last_updated":"2025-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":1,"output":5,"cache_read":0.5}},"qwen/qwen3-next-80b-a3b-thinking":{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"Qwen3 Next 80B A3B (Thinking)","description":"Efficient Qwen thinking model for local reasoning, math, and coding agents","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":0.15,"output":0.65,"cache_read":0.075}},"qwen/qwen3.5-9b":{"id":"qwen/qwen3.5-9b","name":"Qwen3.5 9B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":65536},"cost":{"input":0.05,"output":0.15,"cache_read":0.025}},"qwen/qwen3-coder-flash":{"id":"qwen/qwen3-coder-flash","name":"Qwen3 Coder Flash","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.3,"output":1.5,"cache_read":0.15}},"qwen/qwen3-14b":{"id":"qwen/qwen3-14b","name":"Qwen 3 14b","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":41000,"input":41000,"output":32768},"cost":{"input":0.08,"output":0.24,"cache_read":0.04}},"qwen/Qwen3-8B":{"id":"qwen/Qwen3-8B","name":"Qwen 3 8B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":41000,"input":41000,"output":32768},"cost":{"input":0.47,"output":0.47,"cache_read":0.235}},"qwen/qwen3.8-27b-obliterated":{"id":"qwen/qwen3.8-27b-obliterated","name":"Qwen 3.8 27B Obliterated","description":"Qwen 3.8 27B Obliterated is an open-weight multimodal model LoRA-tuned for fewer refusals across chat, coding, reasoning, tool use, and long-context work.","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-08-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.25,"output":1.5,"cache_read":0.125}},"qwen/qwen3.5-plus-thinking":{"id":"qwen/qwen3.5-plus-thinking","name":"Qwen3.5 Plus Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-02-16","last_updated":"2026-02-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":983616,"input":983616,"output":65536},"cost":{"input":0.4,"output":2.4,"cache_read":0.04}},"qwen/qwen3-32b":{"id":"qwen/qwen3-32b","name":"Qwen 3 32b","description":"Dense open Qwen model for self-hosted chat, reasoning, and coding","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":41000,"input":41000,"output":32768},"cost":{"input":0.1,"output":0.3,"cache_read":0.05}},"qwen/qwen3-coder":{"id":"qwen/qwen3-coder","name":"Qwen 3 Coder 480B","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262000,"input":262000,"output":65536},"cost":{"input":0.13,"output":0.5,"cache_read":0.065}},"qwen/qwen3-coder-next":{"id":"qwen/qwen3-coder-next","name":"Qwen3 Coder Next","description":"Open-weight Qwen coding model for agents, repository edits, and multi-turn tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-09","release_date":"2026-02-03","last_updated":"2026-02-03","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.2,"output":1.5,"cache_read":0.1}},"qwen/qwen3.5-397b-a17b":{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen3.5 397B A17B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":258048,"input":258048,"output":65536},"cost":{"input":0.6,"output":3.6,"cache_read":0.3}},"qwen/Qwen3-235B-A22B-Instruct-2507":{"id":"qwen/Qwen3-235B-A22B-Instruct-2507","name":"Qwen 3 235b A22B 2507","description":"Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-21","last_updated":"2025-07-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":235929},"cost":{"input":0.13,"output":0.5,"cache_read":0.065}},"qwen/qwen3.6-35b-a3b-uncensored:thinking":{"id":"qwen/qwen3.6-35b-a3b-uncensored:thinking","name":"Qwen 3.6 35B A3B Uncensored Thinking","description":"Qwen 3.6 35B A3B Uncensored with thinking enabled for more deliberate coding, multimodal analysis, tool use, and complex chat tasks.","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.15,"output":0.95,"cache_read":0.075}},"qwen/Qwen3-VL-235B-A22B-Instruct":{"id":"qwen/Qwen3-VL-235B-A22B-Instruct","name":"Qwen3 VL 235B A22B Instruct","description":"Qwen vision-language instruct model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.3,"output":1.2,"cache_read":0.15}},"qwen/qwen3.8-27b-fable":{"id":"qwen/qwen3.8-27b-fable","name":"Qwen 3.8 27B Fable","description":"Qwen 3.8 27B Fable is an open-weight multimodal creative finetune for expressive dialogue, long-form storytelling, character work, and roleplay.","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-29","last_updated":"2026-08-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.25,"output":1.5,"cache_read":0.125}},"qwen/qwen3-max":{"id":"qwen/qwen3-max","name":"Qwen3 Max","description":"Flagship Qwen3 model for coding agents, complex reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":1.2002,"output":6.001,"cache_read":0.6001}},"qwen/Qwen3-Next-80B-A3B-Instruct":{"id":"qwen/Qwen3-Next-80B-A3B-Instruct","name":"Qwen3 Next 80B A3B (Instruct)","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":235929},"cost":{"input":0.15,"output":0.65,"cache_read":0.075}},"qwen/qwen3.8-2.4t-a95b":{"id":"qwen/qwen3.8-2.4t-a95b","name":"Qwen3.8 2.4T A95B (Max)","description":"Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":991000,"input":991000,"output":65536},"cost":{"input":2,"output":6,"cache_read":0.25,"cache_write":2.5}},"qwen/qwen3.8-27b-uncensored":{"id":"qwen/qwen3.8-27b-uncensored","name":"Qwen 3.8 27B Uncensored","description":"Qwen 3.8 27B Uncensored is an NVFP4 open-weight multimodal model LoRA-tuned for fewer refusals across chat, coding, reasoning, tool use, and long-context work.","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-08-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.25,"output":1.5,"cache_read":0.125}},"qwen/qwen3-30b-a3b":{"id":"qwen/qwen3-30b-a3b","name":"Qwen3 30B A3B","description":"Sparse MoE Qwen model with 3B active parameters for efficient chat and reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-04-28","last_updated":"2025-04-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":41000,"input":41000,"output":32768},"cost":{"input":0.1,"output":0.3,"cache_read":0.05}},"qwen/Qwen3-235B-A22B-Thinking-2507":{"id":"qwen/Qwen3-235B-A22B-Thinking-2507","name":"Qwen 3 235b A22B 2507 Thinking","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-09-11","last_updated":"2025-09-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":117964},"cost":{"input":0.3,"output":0.5,"cache_read":0.15}},"qwen/qwen3.6-35b-a3b-uncensored":{"id":"qwen/qwen3.6-35b-a3b-uncensored","name":"Qwen 3.6 35B A3B Uncensored","description":"Qwen 3.6 35B A3B Uncensored is an NVFP4 open-weight mixture-of-experts model LoRA-tuned for fewer refusals across chat, coding, tool use, and multimodal tasks.","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.15,"output":0.95,"cache_read":0.075}},"qwen/qwen-2.5-72b-instruct":{"id":"qwen/qwen-2.5-72b-instruct","name":"Qwen2.5 72B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-07-03","last_updated":"2025-07-03","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":8192},"cost":{"input":0.357,"output":0.408,"cache_read":0.1785}},"qwen/qwen3-235b-a22b":{"id":"qwen/qwen3-235b-a22b","name":"Qwen 3 235b A22B","description":"Large open Qwen MoE for multilingual reasoning, coding, and tool use","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.3,"output":0.5,"cache_read":0.15}},"qwen/Qwen3.6-35B-A3B":{"id":"qwen/Qwen3.6-35B-A3B","name":"Qwen3.6 35B A3B","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.112,"output":0.8,"cache_read":0.056}},"qwen/Qwen3.6-35B-A3B:thinking":{"id":"qwen/Qwen3.6-35B-A3B:thinking","name":"Qwen3.6 35B A3B Thinking","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.112,"output":0.8,"cache_read":0.056}},"qwen/qwen3.8-27b-uncensored:thinking":{"id":"qwen/qwen3.8-27b-uncensored:thinking","name":"Qwen 3.8 27B Uncensored Thinking","description":"Qwen 3.8 27B Uncensored with thinking enabled for more deliberate creative work, coding, multimodal analysis, tool use, and long-context problem solving.","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-08-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.25,"output":1.5,"cache_read":0.125}},"qwen/qwen3.5-plus":{"id":"qwen/qwen3.5-plus","name":"Qwen3.5 Plus","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-02-16","last_updated":"2026-02-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":983616,"input":983616,"output":65536},"cost":{"input":0.4,"output":2.4,"cache_read":0.04}},"qwen/qwen3.8-27b-obliterated:thinking":{"id":"qwen/qwen3.8-27b-obliterated:thinking","name":"Qwen 3.8 27B Obliterated Thinking","description":"Qwen 3.8 27B Obliterated with thinking enabled for more deliberate creative work, coding, multimodal analysis, tool use, and long-context problem solving.","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-08-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.25,"output":1.5,"cache_read":0.125}},"qwen/Qwen2.5-Coder-32B-Instruct":{"id":"qwen/Qwen2.5-Coder-32B-Instruct","name":"Qwen 2.5 Coder 32b","description":"Open coding-focused Qwen model for code generation, repair, and repository reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2024-11-12","last_updated":"2024-11-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32000,"input":32000,"output":8192},"cost":{"input":0.2006,"output":0.2006,"cache_read":0.1003}},"MarinaraSpaghetti/NemoMix-Unleashed-12B":{"id":"MarinaraSpaghetti/NemoMix-Unleashed-12B","name":"NemoMix 12B Unleashed","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-nemo","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"aion-labs/aion-2.0":{"id":"aion-labs/aion-2.0","name":"AionLabs: Aion-2.0","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.8,"output":1.6,"cache_read":0.2}},"aion-labs/aion-rp-llama-3.1-8b":{"id":"aion-labs/aion-rp-llama-3.1-8b","name":"Llama 3.1 8b (uncensored)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.8,"output":1.6,"cache_read":0.4}},"aion-labs/aion-3.0":{"id":"aion-labs/aion-3.0","name":"AionLabs: Aion 3.0","description":"Aion 3.0 is a GLM-family collaborative generation model tuned for immersive roleplay and storytelling, with stronger narrative structure, tension, conflict, and nuanced mature themes.","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-07","last_updated":"2026-07-07","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":3,"output":6,"cache_read":0.75}},"aion-labs/aion-3.0-mini":{"id":"aion-labs/aion-3.0-mini","name":"AionLabs: Aion 3.0 Mini","description":"Aion 3.0 Mini is a DeepSeek-family collaborative generation model tuned for immersive roleplay and storytelling, with stronger narrative structure, tension, conflict, and nuanced mature themes.","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-07-07","last_updated":"2026-07-07","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.7,"output":1.4,"cache_read":0.18}},"nothingiisreal/L3.1-70B-Celeste-V0.1-BF16":{"id":"nothingiisreal/L3.1-70B-Celeste-V0.1-BF16","name":"Llama 3.1 70B Celeste v0.1","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"ornith-ai/ornith-1.5-35b-a3b":{"id":"ornith-ai/ornith-1.5-35b-a3b","name":"Ornith 1.5 35B","description":"Ornith 1.5 35B A3B is an open-weight mixture-of-experts model for agentic coding, tool use, image understanding, and long-context work. This variant disables thinking for faster direct responses.","family":"ornith","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-08-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.05}},"ornith-ai/ornith-1.5-35b-a3b:thinking":{"id":"ornith-ai/ornith-1.5-35b-a3b:thinking","name":"Ornith 1.5 35B Thinking","description":"Ornith 1.5 35B A3B is an open-weight mixture-of-experts model for agentic coding, reasoning, tool use, image understanding, and long-context work. This variant enables thinking by default.","family":"ornith","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-08-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.05}},"ornith-ai/ornith-1.5-9b:thinking":{"id":"ornith-ai/ornith-1.5-9b:thinking","name":"Ornith 1.5 9B Thinking","description":"Ornith 1.5 9B with thinking enabled for more deliberate agentic coding, tool use, visual understanding, and long-context work.","family":"ornith","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.2,"cache_read":0.05}},"ornith-ai/ornith-1.5-9b":{"id":"ornith-ai/ornith-1.5-9b","name":"Ornith 1.5 9B","description":"Ornith 1.5 9B is an FP8 dense open-weight model built for agentic coding, tool use, visual understanding, and efficient long-context work. This variant disables thinking for faster direct responses.","family":"ornith","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.2,"cache_read":0.05}},"abliteration-ai/abliterated-model-large":{"id":"abliteration-ai/abliterated-model-large","name":"Abliterated Model Large","description":"Abliteration.ai's large text reasoning model is derived from GLM-5.2 and supports native tool calling, structured output, automatic prompt caching, and a one-million-token context window.","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-27","last_updated":"2026-08-27","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":999990},"cost":{"input":5,"output":5,"cache_read":0.5}},"abliteration-ai/abliterated-model-large-v2":{"id":"abliteration-ai/abliterated-model-large-v2","name":"Abliterated Model Large V2","description":"Abliteration.ai's default large text reasoning model is derived from GLM-5.3 for harder reasoning and evaluation workloads, with automatic prompt caching and a one-million-token context window.","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-31","last_updated":"2026-08-31","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":999990},"cost":{"input":5,"output":5,"cache_read":0.5}},"abliteration-ai/abliterated-model":{"id":"abliteration-ai/abliterated-model","name":"Abliterated Model","description":"Abliteration.ai's multimodal reasoning model supports text and image input, structured output, automatic prompt caching, and a 262K-token context window.","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":false,"structured_output":true,"release_date":"2026-08-27","last_updated":"2026-08-27","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"input":262144,"output":262134},"cost":{"input":3,"output":3,"cache_read":0.3}},"undi95/remm-slerp-l2-13b":{"id":"undi95/remm-slerp-l2-13b","name":"ReMM SLERP 13B","description":"Open Llama multimodal model for image understanding and text reasoning","family":"llama","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-01-01","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":6144,"input":6144,"output":4096},"cost":{"input":0.799,"output":1.207,"cache_read":0.3995}},"chutesai/Mistral-Small-3.2-24B-Instruct-2506":{"id":"chutesai/Mistral-Small-3.2-24B-Instruct-2506","name":"Mistral Small 3.2 24b Instruct","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"chutesai","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-04-15","last_updated":"2025-04-15","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.2,"output":0.4,"cache_read":0.1}},"stepfun-ai/step-3.5-flash":{"id":"stepfun-ai/step-3.5-flash","name":"Step 3.5 Flash","description":"StepFun flash lane for quick multimodal reasoning and coding assistance","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-01-29","last_updated":"2026-02-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.1,"output":0.3,"cache_read":0.05}},"stepfun-ai/step-3.5-flash-2603":{"id":"stepfun-ai/step-3.5-flash-2603","name":"Step 3.5 Flash 2603","description":"StepFun flash model for efficient multimodal reasoning, coding, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.1,"output":0.3,"cache_read":0.05}},"pamanseau/OpenReasoning-Nemotron-32B":{"id":"pamanseau/OpenReasoning-Nemotron-32B","name":"OpenReasoning Nemotron 32B","description":"Nemotron model for efficient reasoning, coding, and specialized AI agents","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2025-08-21","last_updated":"2025-08-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.05}},"deepseek-ai/DeepSeek-V3.1:thinking":{"id":"deepseek-ai/DeepSeek-V3.1:thinking","name":"DeepSeek V3.1 Thinking","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek-thinking","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-08-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.2,"output":0.7,"cache_read":0.1}},"deepseek-ai/deepseek-v3.2-exp-thinking":{"id":"deepseek-ai/deepseek-v3.2-exp-thinking","name":"DeepSeek V3.2 Exp Thinking","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-09-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"input":163840,"output":65536},"cost":{"input":0.28,"output":0.42,"cache_read":0.14}},"deepseek-ai/DeepSeek-V3.1":{"id":"deepseek-ai/DeepSeek-V3.1","name":"DeepSeek V3.1","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.2,"output":0.7,"cache_read":0.1}},"deepseek-ai/DeepSeek-V3.1-Terminus:thinking":{"id":"deepseek-ai/DeepSeek-V3.1-Terminus:thinking","name":"DeepSeek V3.1 Terminus (Thinking)","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek-thinking","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2025-09-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.25,"output":0.7,"cache_read":0.125}},"deepseek-ai/DeepSeek-R1-0528":{"id":"deepseek-ai/DeepSeek-R1-0528","name":"DeepSeek R1 0528","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2025-05-28","last_updated":"2025-05-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"input":163840,"output":32768},"cost":{"input":0.4,"output":1.7,"cache_read":0.2}},"deepseek-ai/deepseek-v3.2-exp":{"id":"deepseek-ai/deepseek-v3.2-exp","name":"DeepSeek V3.2 Exp","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-09-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"input":163840,"output":65536},"cost":{"input":0.28,"output":0.42,"cache_read":0.14}},"deepseek-ai/DeepSeek-V3.1-Terminus":{"id":"deepseek-ai/DeepSeek-V3.1-Terminus","name":"DeepSeek V3.1 Terminus","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2025-08-02","last_updated":"2025-08-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.25,"output":0.7,"cache_read":0.125}},"VongolaChouko/Starcannon-Unleashed-12B-v1.0":{"id":"VongolaChouko/Starcannon-Unleashed-12B-v1.0","name":"Mistral Nemo Starcannon 12b v1","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-nemo","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"poolside/laguna-s-2.1:thinking":{"id":"poolside/laguna-s-2.1:thinking","name":"Laguna S 2.1 Thinking","description":"Agentic coding model from Poolside in the XS size class for local deployment","family":"laguna","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.1,"output":0.2,"cache_read":0.01}},"poolside/laguna-s-2.1":{"id":"poolside/laguna-s-2.1","name":"Laguna S 2.1","description":"Agentic coding model from Poolside in the XS size class for local deployment","family":"laguna","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.1,"output":0.2,"cache_read":0.01}},"featherless-ai/Qwerky-72B":{"id":"featherless-ai/Qwerky-72B","name":"Qwerky 72B","description":"General-purpose chat model for instruction following, writing, and analysis","family":"qwerky","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-03-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32000,"input":32000,"output":8192},"cost":{"input":0.5,"output":0.5,"cache_read":0.25}},"stepfun/step-3.7-flash:thinking":{"id":"stepfun/step-3.7-flash:thinking","name":"Step 3.7 Flash Thinking","description":"Newer StepFun flash model for faster agents, coding, and multimodal prompts","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03-01","release_date":"2026-05-29","last_updated":"2026-05-29","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":256000},"cost":{"input":0.2,"output":1.15,"cache_read":0.04}},"mlabonne/NeuralDaredevil-8B-abliterated":{"id":"mlabonne/NeuralDaredevil-8B-abliterated","name":"Neural Daredevil 8B abliterated","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"input":8192,"output":8192},"cost":{"input":0.44,"output":0.44,"cache_read":0.22}},"mistralai/ministral-14b-2512":{"id":"mistralai/ministral-14b-2512","name":"Ministral 14B","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-04","last_updated":"2025-12-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.2,"output":0.2,"cache_read":0.1}},"mistralai/mistral-large":{"id":"mistralai/mistral-large","name":"Mistral Large 2411","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-02-26","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":102400},"cost":{"input":2.006,"output":6.001,"cache_read":0.2}},"mistralai/codestral-2508":{"id":"mistralai/codestral-2508","name":"Codestral 2508","description":"Mistral coding model for code completion, generation, and developer workflows","family":"codestral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-08-01","last_updated":"2025-08-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":0.3,"output":0.9,"cache_read":0.15}},"mistralai/mistral-large-3-675b-instruct-2512":{"id":"mistralai/mistral-large-3-675b-instruct-2512","name":"Mistral Large 3 675B","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-25","last_updated":"2025-12-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":256000},"cost":{"input":1,"output":3,"cache_read":0.5}},"mistralai/Devstral-Small-2505":{"id":"mistralai/Devstral-Small-2505","name":"Mistral Devstral Small 2505","description":"Mistral coding agent model for repository tasks and software engineering workflows","family":"devstral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-08-02","last_updated":"2025-08-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":8192},"cost":{"input":0.06,"output":0.06,"cache_read":0.03}},"mistralai/devstral-2-123b-instruct-2512":{"id":"mistralai/devstral-2-123b-instruct-2512","name":"Devstral 2 123B","description":"Mistral coding agent model for repository tasks and software engineering workflows","family":"devstral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-09","last_updated":"2025-12-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.4,"output":1.4,"cache_read":0.2}},"mistralai/mistral-saba":{"id":"mistralai/mistral-saba","name":"Mistral Saba","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-02-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":26214},"cost":{"input":0.1989,"output":0.595,"cache_read":0.09945}},"mistralai/ministral-14b-instruct-2512":{"id":"mistralai/ministral-14b-instruct-2512","name":"Ministral 3 14B","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-02","last_updated":"2025-12-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.05}},"mistralai/mistral-small-4-119b-2603:thinking":{"id":"mistralai/mistral-small-4-119b-2603:thinking","name":"Mistral Small 4 119B Thinking","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.4,"output":1.4,"cache_read":0.2}},"mistralai/mistral-small-4-119b-2603":{"id":"mistralai/mistral-small-4-119b-2603","name":"Mistral Small 4 119B","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"release_date":"2026-03-16","last_updated":"2026-03-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.4,"output":1.4,"cache_read":0.2}},"mistralai/ministral-3b-2512":{"id":"mistralai/ministral-3b-2512","name":"Ministral 3B","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-04","last_updated":"2025-12-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.1,"output":0.1,"cache_read":0.05}},"mistralai/mistral-medium-3":{"id":"mistralai/mistral-medium-3","name":"Mistral Medium 3","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.4,"output":2,"cache_read":0.2}},"mistralai/mixtral-8x22b-instruct-v0.1":{"id":"mistralai/mixtral-8x22b-instruct-v0.1","name":"Mixtral 8x22B","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mixtral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":52428},"cost":{"input":2,"output":6,"cache_read":0.2}},"mistralai/Mistral-Nemo-Instruct-2407":{"id":"mistralai/Mistral-Nemo-Instruct-2407","name":"Mistral Nemo","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-nemo","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.1003,"output":0.1207,"cache_read":0.05015}},"mistralai/ministral-8b-2512":{"id":"mistralai/ministral-8b-2512","name":"Ministral 8B","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-04","last_updated":"2025-12-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.15,"output":0.15,"cache_read":0.075}},"mistralai/mistral-medium-3.1":{"id":"mistralai/mistral-medium-3.1","name":"Mistral Medium 3.1","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-05","last_updated":"2025-09-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"input":131072,"output":32768},"cost":{"input":0.4,"output":2,"cache_read":0.2}},"ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.0":{"id":"ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.0","name":"Omega Directive 24B Unslop v2.0","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-08","last_updated":"2025-12-08","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.5,"output":0.5,"cache_read":0.25}},"huihui-ai/DeepSeek-R1-Distill-Llama-70B-abliterated":{"id":"huihui-ai/DeepSeek-R1-Distill-Llama-70B-abliterated","name":"DeepSeek R1 Llama 70B Abliterated","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-01-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.7,"output":0.7,"cache_read":0.35}},"huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated":{"id":"huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated","name":"DeepSeek R1 Qwen Abliterated","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-01-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":1.4,"output":1.4,"cache_read":0.7}},"huihui-ai/Llama-3.3-70B-Instruct-abliterated":{"id":"huihui-ai/Llama-3.3-70B-Instruct-abliterated","name":"Llama 3.3 70B Instruct abliterated","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-08-08","last_updated":"2025-08-08","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.7,"output":0.7,"cache_read":0.35}},"huihui-ai/Qwen2.5-32B-Instruct-abliterated":{"id":"huihui-ai/Qwen2.5-32B-Instruct-abliterated","name":"Qwen 2.5 32B Abliterated","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-01-06","last_updated":"2025-01-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":8192},"cost":{"input":0.7,"output":0.7,"cache_read":0.35}},"xiaomi/mimo-v2.5-pro-crof":{"id":"xiaomi/mimo-v2.5-pro-crof","name":"MiMo V2.5 Pro (Crof)","description":"MiMo V2.5 Pro is Xiaomi's long-context flagship general model for coding and agentic orchestration. This separately served variant is intended for users concerned about censorship on the regular Xiaomi MiMo V2.5 Pro, and it is included in the NanoGPT subscription.","family":"mimo-v2.5-pro","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-07-23","last_updated":"2026-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":131072},"cost":{"input":0.4,"output":0.8,"cache_read":0.003}},"xiaomi/mimo-v2.5:thinking":{"id":"xiaomi/mimo-v2.5:thinking","name":"MiMo V2.5 Thinking","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028,"cache_write":0}},"xiaomi/mimo-v2.5-pro-crof:thinking":{"id":"xiaomi/mimo-v2.5-pro-crof:thinking","name":"MiMo V2.5 Pro Thinking (Crof)","description":"MiMo V2.5 Pro with Xiaomi thinking enabled for coding, long-context reasoning, and agentic orchestration. This separately served thinking variant is intended for users concerned about censorship on the regular Xiaomi MiMo V2.5 Pro, and it is included in the NanoGPT subscription.","family":"mimo-v2.5-pro","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-07-23","last_updated":"2026-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":131072},"cost":{"input":0.4,"output":0.8,"cache_read":0.003}},"xiaomi/mimo-v2.5-pro:thinking":{"id":"xiaomi/mimo-v2.5-pro:thinking","name":"MiMo V2.5 Pro Thinking","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.435,"output":0.87,"cache_read":0.0036,"cache_write":0}},"xiaomi/mimo-v2.5":{"id":"xiaomi/mimo-v2.5","name":"MiMo V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028,"cache_write":0}},"xiaomi/mimo-v2.5-pro":{"id":"xiaomi/mimo-v2.5-pro","name":"MiMo V2.5 Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.435,"output":0.87,"cache_read":0.0036,"cache_write":0}},"shisa-ai/shisa-v2-llama3.3-70b":{"id":"shisa-ai/shisa-v2-llama3.3-70b","name":"Shisa V2 Llama 3.3 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.5,"output":0.5,"cache_read":0.25}},"shisa-ai/shisa-v2.1-llama3.3-70b":{"id":"shisa-ai/shisa-v2.1-llama3.3-70b","name":"Shisa V2.1 Llama 3.3 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":4096},"cost":{"input":0.5,"output":0.5,"cache_read":0.25}},"minimax/minimax-m3:thinking":{"id":"minimax/minimax-m3:thinking","name":"MiniMax M3 Thinking","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-01","last_updated":"2026-06-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":512000,"input":512000,"output":80000},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"minimax/minimax-m2.1":{"id":"minimax/minimax-m2.1","name":"MiniMax M2.1","description":"Earlier MiniMax agent model for practical coding and productivity tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":131072},"cost":{"input":0.33,"output":1.32,"cache_read":0.165}},"minimax/minimax-m2.7":{"id":"minimax/minimax-m2.7","name":"MiniMax M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"input":204800,"output":131072},"cost":{"input":0.315,"output":1.26,"cache_read":0.1575}},"minimax/minimax-m2.7-turbo":{"id":"minimax/minimax-m2.7-turbo","name":"MiniMax M2.7 Turbo","description":"Efficient MiniMax model for quick assistance, coding, and routine automation","family":"minimax-m2.7","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"input":204800,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_read":0.3}},"minimax/minimax-m2.5":{"id":"minimax/minimax-m2.5","name":"MiniMax M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"input":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.15}},"minimax/minimax-m3":{"id":"minimax/minimax-m3","name":"MiniMax M3","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-01","last_updated":"2026-06-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":512000,"input":512000,"output":80000},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"minimax/minimax-m2-her":{"id":"minimax/minimax-m2-her","name":"MiniMax M2-her","description":"MiniMax M2 variant tuned for conversational and character-driven agent interactions","family":"minimax","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-01-23","last_updated":"2026-01-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":65532,"input":65532,"output":2048},"cost":{"input":0.302,"output":1.207,"cache_read":0.151}},"minimax/minimax-01":{"id":"minimax/minimax-01","name":"MiniMax 01","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-01-15","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000192,"input":1000192,"output":16384},"cost":{"input":0.1394,"output":1.122,"cache_read":0.0697}},"minimax/minimax-latest":{"id":"minimax/minimax-latest","name":"MiniMax Latest","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"release_date":"2026-05-03","last_updated":"2026-05-03","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":512000,"input":512000,"output":80000},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"Sao10K/L3-8B-Stheno-v3.2":{"id":"Sao10K/L3-8B-Stheno-v3.2","name":"Sao10K Stheno 8b","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-11-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.2006,"output":0.2006,"cache_read":0.1003}},"Sao10K/L3.3-70B-Euryale-v2.3":{"id":"Sao10K/L3.3-70B-Euryale-v2.3","name":"Llama 3.3 70B Euryale","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":20480,"input":20480,"output":16384},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"Sao10K/L3.1-70B-Euryale-v2.2":{"id":"Sao10K/L3.1-70B-Euryale-v2.2","name":"Llama 3.1 70B Euryale","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":20480,"input":20480,"output":16384},"cost":{"input":0.306,"output":0.357,"cache_read":0.153}},"Sao10K/L3.1-70B-Hanami-x1":{"id":"Sao10K/L3.1-70B-Hanami-x1","name":"Llama 3.1 70B Hanami","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"alibaba/qwen3.6-27b:thinking":{"id":"alibaba/qwen3.6-27b:thinking","name":"Qwen3.6 27B Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":131072}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":260096,"input":260096,"output":65536},"cost":{"input":0.203,"output":2.24,"cache_read":0.1015}},"alibaba/qwen3.6-27b":{"id":"alibaba/qwen3.6-27b","name":"Qwen3.6 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":260096,"input":260096,"output":65536},"cost":{"input":0.203,"output":2.24,"cache_read":0.1015}},"alibaba/qwen3.8-max-0902":{"id":"alibaba/qwen3.8-max-0902","name":"Qwen3.8 Max 0902","description":"2026-09-02 upgraded snapshot of Qwen3.8 Max with stronger coding, collaborative agents, and multimodal document understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.17,"cache_write":2.5}},"alibaba/qwen3.6-flash":{"id":"alibaba/qwen3.6-flash","name":"Qwen3.6 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.6","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-04-27","last_updated":"2026-04-27","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":65536},"cost":{"input":0.19,"output":1.16,"cache_read":0.02,"cache_write":0.24}},"alibaba/qwen3.8-flash":{"id":"alibaba/qwen3.8-flash","name":"Qwen3.8 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991808,"input":991808,"output":131072},"cost":{"input":0.16,"output":0.47,"cache_read":0.016,"cache_write":0.2}},"nvidia/nemotron-3-ultra-550b-a55b:thinking":{"id":"nvidia/nemotron-3-ultra-550b-a55b:thinking","name":"Nvidia Nemotron 3 Ultra 550B Thinking","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-06-04","last_updated":"2026-06-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":0.5,"output":2.5,"cache_read":0.25}},"nvidia/nemotron-3.5-lightning":{"id":"nvidia/nemotron-3.5-lightning","name":"Nvidia Nemotron 3.5 Lightning","description":"Fast NVIDIA Nemotron MoE for reliable agentic tasks across enterprise workloads","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-11","last_updated":"2026-08-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":0.05,"output":0.2,"cache_read":0.01}},"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning":{"id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning","name":"Nvidia Nemotron 3 Nano Omni","description":"Open Nemotron omni model combining reasoning with text, vision, and audio","family":"nemotron","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-28","last_updated":"2026-04-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":65536},"cost":{"input":0.105,"output":0.42,"cache_read":0.0525}},"nvidia/nemotron-3-super-120b-a12b:thinking":{"id":"nvidia/nemotron-3-super-120b-a12b:thinking","name":"Nvidia Nemotron 3 Super 120B Thinking","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-03-01","last_updated":"2026-03-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.05,"output":0.25,"cache_read":0.025}},"nvidia/nemotron-3.5-lightning:thinking":{"id":"nvidia/nemotron-3.5-lightning:thinking","name":"Nvidia Nemotron 3.5 Lightning Thinking","description":"Fast NVIDIA Nemotron MoE for reliable agentic tasks across enterprise workloads","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-11","last_updated":"2026-08-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":0.05,"output":0.2,"cache_read":0.01}},"nvidia/nemotron-3-super-120b-a12b":{"id":"nvidia/nemotron-3-super-120b-a12b","name":"Nvidia Nemotron 3 Super 120B","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-03-01","last_updated":"2026-03-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":16384},"cost":{"input":0.05,"output":0.25,"cache_read":0.025}},"nvidia/Llama-3.1-Nemotron-70B-Instruct-HF":{"id":"nvidia/Llama-3.1-Nemotron-70B-Instruct-HF","name":"Nvidia Nemotron 70b","description":"Nemotron model for efficient reasoning, coding, and specialized AI agents","family":"nemotron","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-04-15","last_updated":"2025-04-15","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.357,"output":0.408,"cache_read":0.1785}},"nvidia/nemotron-3-ultra-550b-a55b":{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"Nvidia Nemotron 3 Ultra 550B","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-06-04","last_updated":"2026-06-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":0.5,"output":2.5,"cache_read":0.25}},"nvidia/Llama-3.3-Nemotron-Super-49B-v1":{"id":"nvidia/Llama-3.3-Nemotron-Super-49B-v1","name":"Nvidia Nemotron Super 49B","description":"Nemotron model for efficient reasoning, coding, and specialized AI agents","family":"nemotron","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-08-08","last_updated":"2025-08-08","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.15,"output":0.15,"cache_read":0.075}},"nvidia/nemotron-3-nano-30b-a3b":{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"Nvidia Nemotron 3 Nano 30B","description":"Small Nemotron 3 MoE for efficient coding, math, and long-context agents","family":"nemotron","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-12-15","last_updated":"2025-12-15","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":235929},"cost":{"input":0.17,"output":0.68,"cache_read":0.085}},"anthropic/claude-opus-4.8":{"id":"anthropic/claude-opus-4.8","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-opus-4.6:thinking":{"id":"anthropic/claude-opus-4.6:thinking","name":"Claude 4.6 Opus Thinking","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-fable-latest":{"id":"anthropic/claude-fable-latest","name":"Claude Fable Latest","description":"Compatibility alias for Claude Fable.","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5}},"anthropic/claude-opus-4.7":{"id":"anthropic/claude-opus-4.7","name":"Claude 4.7 Opus","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-opus-5":{"id":"anthropic/claude-opus-5","name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-sonnet-4.6":{"id":"anthropic/claude-sonnet-4.6","name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":3,"output":15,"cache_read":0.3}},"anthropic/claude-opus-4.6:thinking:low":{"id":"anthropic/claude-opus-4.6:thinking:low","name":"Claude 4.6 Opus Thinking Low","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-opus-latest":{"id":"anthropic/claude-opus-latest","name":"Claude Opus Latest","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-03-29","last_updated":"2026-03-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-opus-4.6:thinking:medium":{"id":"anthropic/claude-opus-4.6:thinking:medium","name":"Claude 4.6 Opus Thinking Medium","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-sonnet-5:thinking":{"id":"anthropic/claude-sonnet-5:thinking","name":"Claude Sonnet 5 Thinking","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"anthropic/claude-opus-4.6":{"id":"anthropic/claude-opus-4.6","name":"Claude 4.6 Opus","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-fable-5":{"id":"anthropic/claude-fable-5","name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"anthropic/claude-haiku-latest":{"id":"anthropic/claude-haiku-latest","name":"Claude Haiku Latest","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024,"max":63999}],"tool_call":true,"structured_output":true,"release_date":"2026-03-29","last_updated":"2026-03-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1}},"anthropic/claude-sonnet-4.6:thinking":{"id":"anthropic/claude-sonnet-4.6:thinking","name":"Claude Sonnet 4.6 Thinking","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":3,"output":15,"cache_read":0.3}},"anthropic/claude-sonnet-latest":{"id":"anthropic/claude-sonnet-latest","name":"Claude Sonnet Latest","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-03-01","last_updated":"2026-03-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"anthropic/claude-opus-4.8:thinking":{"id":"anthropic/claude-opus-4.8:thinking","name":"Claude Opus 4.8 Thinking","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-opus-4.6:thinking:max":{"id":"anthropic/claude-opus-4.6:thinking:max","name":"Claude 4.6 Opus Thinking Max","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-opus-4.7:thinking":{"id":"anthropic/claude-opus-4.7:thinking","name":"Claude 4.7 Opus Thinking","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"anthropic/claude-sonnet-5":{"id":"anthropic/claude-sonnet-5","name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"anthropic/claude-fable-5.1":{"id":"anthropic/claude-fable-5.1","name":"Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5}},"google/gemma-4-26b-a4b-it":{"id":"google/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.12,"output":0.38,"cache_read":0.06}},"google/gemini-pro-latest":{"id":"google/gemini-pro-latest","name":"Gemini Pro Latest","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-03-29","last_updated":"2026-03-29","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":0.375}},"google/gemini-3.5-flash-thinking":{"id":"google/gemini-3.5-flash-thinking","name":"Gemini 3.5 Flash Thinking","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"cache_write":0.083333}},"google/gemini-3-flash-preview-thinking":{"id":"google/gemini-3-flash-preview-thinking","name":"Gemini 3 Flash Thinking","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05}},"google/gemini-3.1-pro-preview-customtools":{"id":"google/gemini-3.1-pro-preview-customtools","name":"Gemini 3.1 Pro (Preview Custom Tools)","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":0.375}},"google/gemini-3.1-pro-preview":{"id":"google/gemini-3.1-pro-preview","name":"Gemini 3.1 Pro (Preview)","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":0.375}},"google/gemini-3.6-flash":{"id":"google/gemini-3.6-flash","name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.041667}},"google/gemini-3.1-flash-lite":{"id":"google/gemini-3.1-flash-lite","name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"cache_write":0.08333}},"google/gemini-3.5-flash":{"id":"google/gemini-3.5-flash","name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"cache_write":0.083333}},"google/gemini-3.1-pro-preview-high":{"id":"google/gemini-3.1-pro-preview-high","name":"Gemini 3.1 Pro (Preview High)","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-02-21","last_updated":"2026-02-21","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":0.375}},"google/gemini-3.5-flash-lite":{"id":"google/gemini-3.5-flash-lite","name":"Gemini 3.5 Flash Lite","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"cache_write":0.08333}},"google/gemma-4-31b-it:thinking":{"id":"google/gemma-4-31b-it:thinking","name":"Gemma 4 31B Thinking","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.1,"output":0.35,"cache_read":0.05}},"google/gemini-flash-lite-latest":{"id":"google/gemini-flash-lite-latest","name":"Gemini Flash Lite Latest","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"cache_write":0.08333}},"google/gemma-4-31b-it":{"id":"google/gemma-4-31b-it","name":"Gemma 4 31B","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.1,"output":0.45,"cache_read":0.05}},"google/gemma-4-26b-a4b-it:thinking":{"id":"google/gemma-4-26b-a4b-it:thinking","name":"Gemma 4 26B A4B Thinking","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.13,"output":0.4,"cache_read":0.065}},"google/gemini-3-flash-preview":{"id":"google/gemini-3-flash-preview","name":"Gemini 3 Flash (Preview)","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05}},"google/gemini-3.8-flash":{"id":"google/gemini-3.8-flash","name":"Gemini 3.8 Flash","description":"Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.075}},"google/gemini-3.7-flash":{"id":"google/gemini-3.7-flash","name":"Gemini 3.7 Flash","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.075}},"google/gemini-flash-latest":{"id":"google/gemini-flash-latest","name":"Gemini Flash Latest","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.075}},"google/gemini-3.1-pro-preview-low":{"id":"google/gemini-3.1-pro-preview-low","name":"Gemini 3.1 Pro (Preview Low)","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-02-21","last_updated":"2026-02-21","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048756,"input":1048756,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":0.375}},"baseten/Kimi-K2-Instruct-FP4":{"id":"baseten/Kimi-K2-Instruct-FP4","name":"Kimi K2 0711 Instruct FP4","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-07-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":131072},"cost":{"input":0.4,"output":1.8,"cache_read":0.2}},"crofai/greg-2-super":{"id":"crofai/greg-2-super","name":"Greg 2 Super","description":"Greg 2 Super is CrofAI's balanced Greg 2 model for strong UI design, frontend iteration, coding, writing, and everyday agent tasks at a lower cost than Ultra.","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-06-19","last_updated":"2026-06-19","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":229376,"input":229376,"output":229376},"cost":{"input":1.5,"output":5,"cache_read":0.25}},"crofai/greg-2-ultra":{"id":"crofai/greg-2-ultra","name":"Greg 2 Ultra","description":"Greg 2 Ultra is CrofAI's most capable Greg 2 model, tuned for premium UI design, agentic coding, creative writing, and higher-end general reasoning tasks.","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-06-19","last_updated":"2026-06-19","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":229376,"input":229376,"output":229376},"cost":{"input":3,"output":10,"cache_read":0.5}},"nex-agi/nex-n2-mini":{"id":"nex-agi/nex-n2-mini","name":"Nex N2 Mini","description":"Nex AGI's open-source agentic mixture-of-experts model in the Nex N2 family. It accepts text and image input and is built for coding, tool use, structured outputs, and optional reasoning with a 256K context window.","family":"agi","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"release_date":"2026-06-24","last_updated":"2026-06-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":235929},"cost":{"input":0.025,"output":0.1,"cache_read":0.0025}},"nex-agi/nex-n2-pro":{"id":"nex-agi/nex-n2-pro","name":"Nex N2 Pro","description":"Nex AGI's open-source agentic reasoning model, post-trained on Qwen3.5-397B-A17B. It is built for agentic coding, software engineering, deep research, tool use, and long-horizon tasks with a 256K context window.","family":"agi","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"release_date":"2026-06-04","last_updated":"2026-06-04","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":235929},"cost":{"input":0.5,"output":2.5,"cache_read":0.25}},"thinkingmachines/Inkling-Small":{"id":"thinkingmachines/Inkling-Small","name":"Inkling Small","description":"Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio","family":"ling","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-07-30","last_updated":"2026-07-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":524288,"input":524288,"output":32768},"cost":{"input":0.5,"output":1.2,"cache_read":0.1}},"thinkingmachines/inkling:thinking":{"id":"thinkingmachines/inkling:thinking","name":"Inkling Thinking","description":"Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio","family":"ling","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":true,"limit":{"context":1048000,"input":1048000,"output":32768},"cost":{"input":1,"output":4.05,"cache_read":0.17}},"thinkingmachines/inkling":{"id":"thinkingmachines/inkling","name":"Inkling","description":"Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio","family":"ling","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":true,"limit":{"context":1048000,"input":1048000,"output":32768},"cost":{"input":1,"output":4.05,"cache_read":0.17}},"thinkingmachines/Inkling-Small:thinking":{"id":"thinkingmachines/Inkling-Small:thinking","name":"Inkling Small Thinking","description":"Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio","family":"ling","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-07-30","last_updated":"2026-07-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":524288,"input":524288,"output":32768},"cost":{"input":0.5,"output":1.2,"cache_read":0.1}},"meta/muse-spark-1.3":{"id":"meta/muse-spark-1.3","name":"Muse Spark 1.3","description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":943718},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"meta/muse-spark-1.2":{"id":"meta/muse-spark-1.2","name":"Muse Spark 1.2","description":"Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-05","last_updated":"2026-08-05","modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"meta/muse-spark-1.2-contributor":{"id":"meta/muse-spark-1.2-contributor","name":"Muse Spark 1.2 Contributor (Data Used for Training)","description":"A much cheaper opt-in version of Muse Spark 1.2 with the same multimodal coding and agentic capabilities. Prompts and outputs sent to this Contributor model may be used by Meta for training and to improve its products; use the standard Muse Spark 1.2 model if you do not want your data used for training.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-05","last_updated":"2026-08-05","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":0.1,"output":0.2,"cache_read":0.002}},"meta/muse-spark-1.3-contributor":{"id":"meta/muse-spark-1.3-contributor","name":"Muse Spark 1.3 Contributor","description":"Meta's Muse Spark 1.3 Contributor is a frontier multimodal reasoning model for long-horizon coding and agentic workflows, with strong gains in computer use, browsing, professional tool use, codebase understanding, and million-token retrieval. It accepts text, images, audio, video, and files, supports tool calling and structured output, and always reasons before answering. Prompts and outputs may be used by Meta for training and to improve its products.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":943718},"cost":{"input":0.1,"output":0.2,"cache_read":0.002}},"meta/muse-spark-1.1":{"id":"meta/muse-spark-1.1","name":"Muse Spark 1.1","description":"Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-04-08","last_updated":"2026-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"meta/muse-glimmer-30b":{"id":"meta/muse-glimmer-30b","name":"Muse Glimmer 30B","description":"Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-01-04","release_date":"2026-08-10","last_updated":"2026-08-10","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":117964},"cost":{"input":0.35,"output":1.5,"cache_read":0.04}},"perceptron/perceptron-mk1":{"id":"perceptron/perceptron-mk1","name":"Perceptron Mk1","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":true,"release_date":"2026-05-12","last_updated":"2026-05-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"input":32768,"output":8192},"cost":{"input":0.15,"output":1.5,"cache_read":0.075}},"Steelskull/L3.3-Cu-Mai-R1-70b":{"id":"Steelskull/L3.3-Cu-Mai-R1-70b","name":"Llama 3.3 70B Cu Mai","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"Steelskull/L3.3-Electra-R1-70b":{"id":"Steelskull/L3.3-Electra-R1-70b","name":"Steelskull Electra R1 70b","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.69989,"output":0.69989,"cache_read":0.349945}},"Steelskull/L3.3-Nevoria-R1-70b":{"id":"Steelskull/L3.3-Nevoria-R1-70b","name":"Steelskull Nevoria R1 70b","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"Steelskull/L3.3-MS-Nevoria-70b":{"id":"Steelskull/L3.3-MS-Nevoria-70b","name":"Steelskull Nevoria 70b","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"bytedance/doubao-seed-2.1-turbo":{"id":"bytedance/doubao-seed-2.1-turbo","name":"Doubao Seed 2.1 Turbo","description":"Fast, lower-cost model in the Doubao Seed 2.1 family for everyday chat, coding assistance, document work, and high-throughput productivity tasks. Supports a 256k context window and up to 128k output tokens. Note: privacy and logging guarantees may be limited.","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-06-23","last_updated":"2026-06-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":128000},"cost":{"input":0.5,"output":2.5,"cache_read":0.25}},"bytedance/doubao-seed-2.1-pro":{"id":"bytedance/doubao-seed-2.1-pro","name":"Doubao Seed 2.1 Pro","description":"Higher-capability model in the Doubao Seed 2.1 family for agentic coding, long-context analysis, complex instruction following, and productivity workflows. Supports a 256k context window and up to 128k output tokens. Note: privacy and logging guarantees may be limited.","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-06-23","last_updated":"2026-06-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.5}},"bytedance/doubao-seed-character":{"id":"bytedance/doubao-seed-character","name":"Doubao Seed Character","description":"ByteDance's character-focused Doubao Seed model for roleplay, persona consistency, dialogue, and creative character interactions. It supports text and image input with a 128k context window. Requests route through ZenMux to ByteDance; ZenMux does not publish a model-API zero-retention or training guarantee, so avoid sensitive data.","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":false,"structured_output":false,"release_date":"2026-07-18","last_updated":"2026-07-18","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":32768},"cost":{"input":0.1179,"output":0.2947,"cache_read":0.0236,"cache_write":0.0025}},"bytedance-seed/seed-2-1-turbo":{"id":"bytedance-seed/seed-2-1-turbo","name":"ByteDance Seed 2.1 Turbo","description":"ByteDance Seed 2.1 Turbo is a multimodal model for coding and long-horizon agent workflows, including end-to-end software delivery and multi-step task execution. It supports text, image, and video input with a 262k context window.","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"input":262144,"output":235929},"cost":{"input":0.5,"output":2.5,"cache_read":0.25}},"bytedance-seed/seed-2.0-code":{"id":"bytedance-seed/seed-2.0-code","name":"ByteDance Seed 2.0 Code","description":"ByteDance Seed coding model for multimodal software engineering and long-running agents","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.5,"output":3,"cache_read":0.25}},"bytedance-seed/seed-2.0-lite":{"id":"bytedance-seed/seed-2.0-lite","name":"ByteDance Seed 2.0 Lite","description":"Cost-efficient ByteDance Seed 2.0 model for production chat, analysis, and structured generation","family":"seed","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.25,"output":2,"cache_read":0.125}},"NeverSleep/Lumimaid-v0.2-70B":{"id":"NeverSleep/Lumimaid-v0.2-70B","name":"Lumimaid v0.2","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":1,"output":1.5,"cache_read":0.5}},"TEE/qwen3.5-27b":{"id":"TEE/qwen3.5-27b","name":"Qwen3.5 27B TEE","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.3,"output":2.4,"cache_read":0.15}},"TEE/qwen3.8-27b":{"id":"TEE/qwen3.8-27b","name":"Qwen3.8 27B TEE","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":262144},"cost":{"input":0.4,"output":3,"cache_read":0.15}},"TEE/kimi-k2.6":{"id":"TEE/kimi-k2.6","name":"Kimi K2.6 TEE","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":1.5,"output":5.25,"cache_read":0.375}},"TEE/glm-5.2":{"id":"TEE/glm-5.2","name":"GLM 5.2 TEE","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1.4,"output":4.6,"cache_read":0.5}},"TEE/gemma4-31b":{"id":"TEE/gemma4-31b","name":"Gemma 4 31B","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-04-04","last_updated":"2026-04-04","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.4,"output":1,"cache_read":0.4}},"TEE/deepseek-v4-flash":{"id":"TEE/deepseek-v4-flash","name":"DeepSeek V4 Flash TEE","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":393216},"cost":{"input":0.2,"output":0.4,"cache_read":0.04}},"TEE/kimi-k2.7-code":{"id":"TEE/kimi-k2.7-code","name":"Kimi K2.7 Code TEE","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.95,"output":4,"cache_read":0.19}},"TEE/gpt-oss-20b":{"id":"TEE/gpt-oss-20b","name":"GPT-OSS 20B TEE","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":8192},"cost":{"input":0.2,"output":0.8,"cache_read":0.1}},"TEE/qwen2.5-vl-72b-instruct":{"id":"TEE/qwen2.5-vl-72b-instruct","name":"Qwen2.5 VL 72B TEE","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-02-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":8192},"cost":{"input":0.7,"output":0.7,"cache_read":0.35}},"TEE/gemma4-31b:thinking":{"id":"TEE/gemma4-31b:thinking","name":"Gemma 4 31B Thinking TEE","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-05-02","last_updated":"2026-05-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":131072},"cost":{"input":0.4,"output":1,"cache_read":0.4}},"TEE/qwen3.5-397b-a17b":{"id":"TEE/qwen3.5-397b-a17b","name":"Qwen3.5 397B A17B TEE","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.55,"output":3.5,"cache_read":0.275}},"TEE/gemma-4-26b-a4b-uncensored":{"id":"TEE/gemma-4-26b-a4b-uncensored","name":"Gemma 4 26B A4B Uncensored TEE","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2026-05-23","last_updated":"2026-05-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":65536},"cost":{"input":0.15,"output":0.7,"cache_read":0.075}},"TEE/qwen3.6-27b":{"id":"TEE/qwen3.6-27b","name":"Qwen3.6 27B TEE","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.32,"output":2.7,"cache_read":0.16}},"TEE/kimi-k3":{"id":"TEE/kimi-k3","name":"Kimi K3 TEE","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":65535},"cost":{"input":3,"output":15,"cache_read":1.5}},"TEE/deepseek-v3.2":{"id":"TEE/deepseek-v3.2","name":"DeepSeek V3.2 TEE","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":164000,"input":164000,"output":65536},"cost":{"input":0.5,"output":1,"cache_read":0.25}},"TEE/qwen3.6-35b-a3b":{"id":"TEE/qwen3.6-35b-a3b","name":"Qwen3.6 35B A3B TEE","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":262144},"cost":{"input":0.2,"output":1.27,"cache_read":0.1}},"TEE/glm-5.3-flash":{"id":"TEE/glm-5.3-flash","name":"GLM 5.3 Flash TEE","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03}},"TEE/muse-glimmer-30b":{"id":"TEE/muse-glimmer-30b","name":"Muse Glimmer 30B TEE","description":"Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-01-04","release_date":"2026-08-10","last_updated":"2026-08-10","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":131072},"cost":{"input":0.35,"output":1.5,"cache_read":0.04}},"TEE/gemma-4-31b-it":{"id":"TEE/gemma-4-31b-it","name":"Gemma 4 31B IT TEE","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":262144},"cost":{"input":0.15,"output":0.46,"cache_read":0.075}},"TEE/qwen3.6-35b-a3b-uncensored":{"id":"TEE/qwen3.6-35b-a3b-uncensored","name":"Qwen3.6 35B A3B Uncensored TEE","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024,"max":131072}],"tool_call":true,"structured_output":true,"release_date":"2026-05-23","last_updated":"2026-05-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":131072},"cost":{"input":0.3,"output":1.5,"cache_read":0.15}},"TEE/qwen3-8b":{"id":"TEE/qwen3-8b","name":"Qwen3 8B TEE","description":"Qwen3 8B is an open-weight dense language model from Alibaba's Qwen team for efficient dialogue, reasoning, mathematics, coding, and tool use. Running inside a TEE (Trusted Execution Environment), with provider attestation support.","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-09-04","last_updated":"2026-09-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":40960,"input":40960,"output":8192},"cost":{"input":0.11,"output":0.45,"cache_read":0.11}},"TEE/glm-5.1":{"id":"TEE/glm-5.1","name":"GLM 5.1 TEE","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"input":202752,"output":65535},"cost":{"input":1.5,"output":5.25,"cache_read":0.3}},"TEE/glm-5.2:thinking":{"id":"TEE/glm-5.2:thinking","name":"GLM 5.2 Thinking TEE","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1.4,"output":4.6,"cache_read":0.5}},"TEE/gpt-oss-120b":{"id":"TEE/gpt-oss-120b","name":"GPT-OSS 120B TEE","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":2,"output":2,"cache_read":2}},"TEE/llama3-3-70b":{"id":"TEE/llama3-3-70b","name":"Llama 3.3 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-03","last_updated":"2025-07-03","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":1.75,"output":2.75,"cache_read":1.75}},"TEE/glm-5.1-thinking":{"id":"TEE/glm-5.1-thinking","name":"GLM 5.1 Thinking TEE","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"input":202752,"output":65535},"cost":{"input":1.5,"output":5.25,"cache_read":0.3}},"TEE/glm-5.3":{"id":"TEE/glm-5.3","name":"GLM 5.3 TEE","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"meganova-ai/manta-mini-1.0":{"id":"meganova-ai/manta-mini-1.0","name":"Manta Mini 1.0","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-20","last_updated":"2025-12-20","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"input":8192,"output":8192},"cost":{"input":0.02,"output":0.16,"cache_read":0.01}},"meganova-ai/manta-flash-1.0":{"id":"meganova-ai/manta-flash-1.0","name":"Manta Flash 1.0","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-20","last_updated":"2025-12-20","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":16384,"input":16384,"output":16384},"cost":{"input":0.02,"output":0.16,"cache_read":0.01}},"meganova-ai/manta-pro-1.0":{"id":"meganova-ai/manta-pro-1.0","name":"Manta Pro 1.0","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-20","last_updated":"2025-12-20","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":65536,"input":65536,"output":32768},"cost":{"input":0.06,"output":0.5,"cache_read":0.03}},"inception/mercury-2.5-preview":{"id":"inception/mercury-2.5-preview","name":"Mercury 2.5 Preview","description":"Mercury 2.5 Preview is Inception's latest and most intelligent diffusion language model. Instead of generating tokens strictly one at a time, it produces and refines multiple tokens in parallel, reaching up to 1,107 tokens per second on standard GPUs. It delivers a 10+ point intelligence gain over Mercury 2, with tunable reasoning, parallel tool calls, schema-aligned JSON output, and a 260K context window. It is built for latency-sensitive production work such as search agents, voice pipelines, customer support, rapid coding iteration, and coding subagents.","family":"mercury","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-09-01","last_updated":"2026-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":260000,"input":260000,"output":65536},"cost":{"input":0.04,"output":0.15,"cache_read":0.004}},"unsloth/gemma-3-4b-it":{"id":"unsloth/gemma-3-4b-it","name":"Gemma 3 4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"unsloth","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-03-10","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.2006,"output":0.2006,"cache_read":0.1003}},"unsloth/gemma-3-27b-it":{"id":"unsloth/gemma-3-27b-it","name":"Gemma 3 27B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"unsloth","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-03-10","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":96000},"cost":{"input":0.2992,"output":0.2992,"cache_read":0.1496}},"unsloth/gemma-3-12b-it":{"id":"unsloth/gemma-3-12b-it","name":"Gemma 3 12B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"unsloth","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-03-10","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":0.272,"output":0.272,"cache_read":0.136}},"microsoft/wizardlm-2-8x22b":{"id":"microsoft/wizardlm-2-8x22b","name":"WizardLM-2 8x22B","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-04-15","last_updated":"2025-04-15","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"inflatebot/MN-12B-Mag-Mell-R1":{"id":"inflatebot/MN-12B-Mag-Mell-R1","name":"Mag Mell R1","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-nemo","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"deepcogito/cogito-v1-preview-qwen-32B":{"id":"deepcogito/cogito-v1-preview-qwen-32B","name":"Cogito v1 Preview Qwen 32B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-05-10","last_updated":"2025-05-10","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":32768},"cost":{"input":1.8,"output":1.8,"cache_read":0.9}},"sakana/fugu-ultra":{"id":"sakana/fugu-ultra","name":"Fugu Ultra","description":"Quality-first multi-agent model for hard research, analysis, and competitions","family":"fugu","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-06-15","last_updated":"2026-06-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":16384},"cost":{"input":5.25,"output":31.5,"cache_read":0.525}},"sakana/fugu-ultra-v1.1":{"id":"sakana/fugu-ultra-v1.1","name":"Fugu Ultra v1.1","description":"Sakana AI's upgraded Fugu Ultra release with stronger coding, agentic task execution, and advanced reasoning through dynamic orchestration of frontier models.","family":"fugu","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":16384},"cost":{"input":5.25,"output":31.5,"cache_read":0.525}},"Envoid/Llama-3.05-Nemotron-Tenyxchat-Storybreaker-70B":{"id":"Envoid/Llama-3.05-Nemotron-Tenyxchat-Storybreaker-70B","name":"Nemotron Tenyxchat Storybreaker 70b","description":"Nemotron model for efficient reasoning, coding, and specialized AI agents","family":"nemotron","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"Envoid/Llama-3.05-NT-Storybreaker-Ministral-70B":{"id":"Envoid/Llama-3.05-NT-Storybreaker-Ministral-70B","name":"Llama 3.05 Storybreaker Ministral 70b","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"NousResearch/hermes-3-llama-3.1-70b":{"id":"NousResearch/hermes-3-llama-3.1-70b","name":"Hermes 3 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"nousresearch","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2026-01-07","last_updated":"2026-01-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":8192},"cost":{"input":0.408,"output":0.408,"cache_read":0.204}},"NousResearch/Hermes-4-70B:thinking":{"id":"NousResearch/Hermes-4-70B:thinking","name":"Hermes 4 (Thinking)","description":"General-purpose chat model for instruction following, writing, and analysis","family":"nousresearch","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-17","last_updated":"2025-09-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.2006,"output":0.3995,"cache_read":0.1003}},"NousResearch/hermes-4-405b":{"id":"NousResearch/hermes-4-405b","name":"Hermes 4 Large","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nousresearch","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"release_date":"2025-08-26","last_updated":"2025-08-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.3,"output":1.2,"cache_read":0.15}},"NousResearch/hermes-4-405b:thinking":{"id":"NousResearch/hermes-4-405b:thinking","name":"Hermes 4 Large (Thinking)","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nousresearch","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.3,"output":1.2,"cache_read":0.15}},"failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5":{"id":"failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5","name":"Llama 3 70B abliterated","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"input":8192,"output":8192},"cost":{"input":0.7,"output":0.7,"cache_read":0.35}},"GalrionSoftworks/MN-LooseCannon-12B-v1":{"id":"GalrionSoftworks/MN-LooseCannon-12B-v1","name":"MN-LooseCannon-12B-v1","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-nemo","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"ibm-granite/granite-4.2-8b":{"id":"ibm-granite/granite-4.2-8b","name":"Granite 4.2 8B","description":"IBM Granite 4.2 8B is an Apache 2.0-licensed dense model with native step-by-step reasoning and specialized training for agentic work. It can plan before acting, sequence tools, navigate codebases, work in terminals, and verify results across coding, search, mathematics, science, and complex instruction-following tasks.","family":"granite","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-31","last_updated":"2026-08-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":117964},"cost":{"input":0.1,"output":0.15,"cache_read":0.05}},"deepseek/deepseek-v4-flash-vision-exp":{"id":"deepseek/deepseek-v4-flash-vision-exp","name":"DeepSeek V4 Flash Vision Exp","description":"Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work","family":"deepseek-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-21","last_updated":"2026-08-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":0.22,"output":0.66,"cache_read":0.007}},"deepseek/deepseek-v4-pro-0813":{"id":"deepseek/deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":1.1,"output":2.5,"cache_read":0.04}},"deepseek/deepseek-v4-flash-0731":{"id":"deepseek/deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.014}},"deepseek/deepseek-v4-flash-latest":{"id":"deepseek/deepseek-v4-flash-latest","name":"DeepSeek V4 Flash Latest","description":"Compatibility alias that routes to the newest dated DeepSeek V4 Flash release. Currently routes to DeepSeek V4 Flash 0731. ⚠️ This route goes directly to DeepSeek, so privacy and logging guarantees are limited.","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-02","last_updated":"2026-08-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.014}},"deepseek/deepseek-v4-flash":{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028}},"deepseek/deepseek-v4-flash-0731:thinking":{"id":"deepseek/deepseek-v4-flash-0731:thinking","name":"DeepSeek V4 Flash 0731 (Thinking)","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"input":1000000,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.014}},"deepseek/deepseek-v4-flash:thinking":{"id":"deepseek/deepseek-v4-flash:thinking","name":"DeepSeek V4 Flash (Thinking)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028}},"deepseek/deepseek-v4-pro-0813:thinking":{"id":"deepseek/deepseek-v4-pro-0813:thinking","name":"DeepSeek V4 Pro 0813 Thinking","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":1.1,"output":2.5,"cache_read":0.04}},"deepseek/deepseek-v4-pro:thinking":{"id":"deepseek/deepseek-v4-pro:thinking","name":"DeepSeek V4 Pro (Thinking)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":1.1,"output":2.2,"cache_read":0.11}},"deepseek/deepseek-v3.2":{"id":"deepseek/deepseek-v3.2","name":"DeepSeek V3.2","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":163000,"input":163000,"output":65536},"cost":{"input":0.28,"output":0.42,"cache_read":0.14}},"deepseek/deepseek-latest":{"id":"deepseek/deepseek-latest","name":"DeepSeek Latest","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-03","last_updated":"2026-05-03","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":1.1,"output":2.5,"cache_read":0.04}},"deepseek/deepseek-v4-pro":{"id":"deepseek/deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":384000},"cost":{"input":1.1,"output":2.2,"cache_read":0.11}},"deepseek/deepseek-v3.2:thinking":{"id":"deepseek/deepseek-v3.2:thinking","name":"DeepSeek V3.2 Thinking","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":163000,"input":163000,"output":65536},"cost":{"input":0.28,"output":0.42,"cache_read":0.14}},"EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.0":{"id":"EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.0","name":"EVA Llama 3.33 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":2.006,"output":2.006,"cache_read":1.003}},"EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1":{"id":"EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1","name":"EVA-LLaMA-3.33-70B-v0.1","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":16384},"cost":{"input":2.006,"output":2.006,"cache_read":1.003}},"EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2":{"id":"EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2","name":"EVA-Qwen2.5-32B-v0.2","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.799,"output":0.799,"cache_read":0.3995}},"EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2":{"id":"EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2","name":"EVA-Qwen2.5-72B-v0.2","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.799,"output":0.799,"cache_read":0.3995}},"amazon/nova-2-lite-v1":{"id":"amazon/nova-2-lite-v1","name":"Amazon Nova 2 Lite","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-12-03","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":65535},"cost":{"input":0.51,"output":4.25,"cache_read":0.255}},"amazon/nova-pro-v1":{"id":"amazon/nova-pro-v1","name":"Amazon Nova Pro 1.0","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova-pro","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-12-03","last_updated":"2024-12-03","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":300000,"input":300000,"output":32000},"cost":{"input":0.799,"output":3.196,"cache_read":0.3995}},"amazon/nova-lite-v1":{"id":"amazon/nova-lite-v1","name":"Amazon Nova Lite 1.0","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-lite","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-12-03","last_updated":"2024-12-03","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":300000,"input":300000,"output":5120},"cost":{"input":0.0595,"output":0.238,"cache_read":0.02975}},"LLM360/K2-Think":{"id":"LLM360/K2-Think","name":"K2-Think","description":"Kimi reasoning model for long-horizon research, planning, and tool use","family":"kimi-thinking","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":32768},"cost":{"input":0.17,"output":0.68,"cache_read":0.085}},"inclusionai/ling-3.0-flash":{"id":"inclusionai/ling-3.0-flash","name":"Ling 3.0 Flash","description":"Ling-3.0-flash is a 124B-parameter Mixture-of-Experts model with approximately 5.1B parameters active per token. It prioritizes token efficiency and production-scale agentic inference, helping coding and tool-using agents complete more work within constrained latency and serving budgets.","family":"ling","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"release_date":"2026-07-23","last_updated":"2026-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.075,"output":0.22,"cache_read":0.015}},"inclusionai/ling-3.0-flash:thinking":{"id":"inclusionai/ling-3.0-flash:thinking","name":"Ling 3.0 Flash Thinking","description":"Ling-3.0-flash Thinking enables visible reasoning on inclusionAI's token-efficient 124B-parameter Mixture-of-Experts model for harder coding, tool use, planning, and production-scale agent workflows.","family":"ling","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"release_date":"2026-07-23","last_updated":"2026-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"input":262144,"output":32768},"cost":{"input":0.075,"output":0.22,"cache_read":0.015}},"MiniMaxAI/MiniMax-M1-80k":{"id":"MiniMaxAI/MiniMax-M1-80k","name":"MiniMax M1 80K","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-01-08","last_updated":"2025-06-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":131072},"cost":{"input":0.6052,"output":2.4225,"cache_read":0.3026}},"anthracite-org/magnum-v4-72b":{"id":"anthracite-org/magnum-v4-72b","name":"Magnum v4 72B","description":"Open Llama multimodal model for image understanding and text reasoning","family":"llama","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-01-01","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":2.006,"output":2.992,"cache_read":1.003}},"anthracite-org/magnum-v2-72b":{"id":"anthracite-org/magnum-v2-72b","name":"Magnum V2 72B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":2.006,"output":2.992,"cache_read":1.003}},"Salesforce/Llama-xLAM-2-70b-fc-r":{"id":"Salesforce/Llama-xLAM-2-70b-fc-r","name":"Llama-xLAM-2 70B fc-r","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-04-13","last_updated":"2025-04-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":2.5,"output":2.5,"cache_read":1.25}},"x-ai/grok-4.20-multi-agent":{"id":"x-ai/grok-4.20-multi-agent","name":"Grok 4.20 Multi-Agent","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"release_date":"2026-03-31","last_updated":"2026-03-31","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"input":2000000,"output":131072},"cost":{"input":2,"output":6,"cache_read":1}},"x-ai/grok-4.3":{"id":"x-ai/grok-4.3","name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":900000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2}},"x-ai/grok-4.20":{"id":"x-ai/grok-4.20","name":"Grok 4.20","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-03-31","last_updated":"2026-03-31","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"input":2000000,"output":131072},"cost":{"input":2,"output":6,"cache_read":1}},"x-ai/grok-latest":{"id":"x-ai/grok-latest","name":"Grok Latest","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-03","last_updated":"2026-05-03","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"input":500000,"output":450000},"cost":{"input":2,"output":6,"cache_read":0.5}},"x-ai/grok-4.5":{"id":"x-ai/grok-4.5","name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-08","last_updated":"2026-07-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"input":500000,"output":450000},"cost":{"input":2,"output":6,"cache_read":0.5}},"x-ai/grok-build-0.1":{"id":"x-ai/grok-build-0.1","name":"Grok Build 0.1","description":"Fast Grok coding model tuned for agentic engineering and iterative edits","family":"grok-build","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"input":256000,"output":230400},"cost":{"input":1,"output":2,"cache_read":0.2}},"x-ai/grok-4.6":{"id":"x-ai/grok-4.6","name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"input":500000,"output":450000},"cost":{"input":2,"output":6,"cache_read":0.5}},"meta-llama/llama-3.1-8b-instruct":{"id":"meta-llama/llama-3.1-8b-instruct","name":"Llama 3.1 8b Instruct","description":"Compact open Llama model for lightweight chat, drafting, and self-hosting","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":0.0544,"output":0.085,"cache_read":0.0272}},"meta-llama/llama-3.2-3b-instruct":{"id":"meta-llama/llama-3.2-3b-instruct","name":"Llama 3.2 3b Instruct","description":"Open Llama multimodal model for image understanding and text reasoning","family":"llama","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-09-25","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":8192},"cost":{"input":0.0306,"output":0.0493,"cache_read":0.0153}},"meta-llama/llama-4-maverick":{"id":"meta-llama/llama-4-maverick","name":"Llama 4 Maverick","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2025-09-05","last_updated":"2025-09-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":65536},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"meta-llama/llama-4-scout":{"id":"meta-llama/llama-4-scout","name":"Llama 4 Scout","description":"Open multimodal Llama model for long-context analysis and efficient agents","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2025-09-05","last_updated":"2025-09-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":328000,"input":328000,"output":65536},"cost":{"input":0.085,"output":0.46,"cache_read":0.0425}},"meta-llama/llama-3.3-70b-instruct":{"id":"meta-llama/llama-3.3-70b-instruct","name":"Llama 3.3 70b Instruct","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":0.05,"output":0.23,"cache_read":0.025}},"abacusai/Dracarys-72B-Instruct":{"id":"abacusai/Dracarys-72B-Instruct","name":"Llama 3.1 70B Dracarys 2","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-08-02","last_updated":"2025-08-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":8192},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"Gryphe/MythoMax-L2-13b":{"id":"Gryphe/MythoMax-L2-13b","name":"MythoMax 13B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-08-08","last_updated":"2025-08-08","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":4096,"input":4096,"output":3686},"cost":{"input":0.1003,"output":0.1003,"cache_read":0.05015}},"openai/o4-mini-high":{"id":"openai/o4-mini-high","name":"OpenAI o4-mini high","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o-mini","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2025-12-04","last_updated":"2025-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai/gpt-5-nano":{"id":"openai/gpt-5-nano","name":"GPT 5 Nano","description":"Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005}},"openai/o3-mini-low":{"id":"openai/o3-mini-low","name":"OpenAI o3-mini (Low)","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o-mini","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-01-31","last_updated":"2025-01-31","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai/o3-pro-2025-06-10":{"id":"openai/o3-pro-2025-06-10","name":"OpenAI o3-pro (2025-06-10)","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o-pro","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2025-06-10","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":22,"output":88,"cache_read":11}},"openai/gpt-4.1-nano":{"id":"openai/gpt-4.1-nano","name":"GPT 4.1 Nano","description":"Tiny GPT-4.1 option for classification, routing, and very high-volume tasks","family":"gpt-nano","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"input":1047576,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.025}},"openai/gpt-5-pro":{"id":"openai/gpt-5-pro","name":"GPT 5 Pro","description":"Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high"]}],"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-10-06","last_updated":"2025-10-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":15,"output":120,"cache_read":1.5}},"openai/o3-mini-high":{"id":"openai/o3-mini-high","name":"OpenAI o3-mini (High)","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o-mini","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2025-01-31","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai/gpt-5.1-codex-mini":{"id":"openai/gpt-5.1-codex-mini","name":"GPT 5.1 Codex Mini","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"openai/gpt-6-astra-pro":{"id":"openai/gpt-6-astra-pro","name":"GPT 6 Astra Pro","description":"GPT 6 Astra in Pro reasoning mode. Uses additional model work for difficult tasks, with higher latency and token usage at the same per-token rates. Reasoning effort remains independently configurable.","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-09-04","last_updated":"2026-09-04","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"openai/gpt-5.1-codex":{"id":"openai/gpt-5.1-codex","name":"GPT 5.1 Codex","description":"Codex GPT for repository edits, code review, and practical software agents","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/gpt-5.6-sol":{"id":"openai/gpt-5.6-sol","name":"GPT 5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"openai/gpt-4o-2024-08-06":{"id":"openai/gpt-4o-2024-08-06","name":"GPT-4o (2024-08-06)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-09","release_date":"2024-08-06","last_updated":"2024-08-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"openai/gpt-5.2-codex":{"id":"openai/gpt-5.2-codex","name":"GPT 5.2 Codex","description":"Code-specialist GPT for repository edits, reviews, and long-running software agents","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-5.1-2025-11-13":{"id":"openai/gpt-5.1-2025-11-13","name":"GPT-5.1 (2025-11-13)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2025-11-13","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":32768},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/gpt-latest":{"id":"openai/gpt-latest","name":"GPT Latest","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-03-29","last_updated":"2026-03-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"openai/gpt-6-astra":{"id":"openai/gpt-6-astra","name":"GPT 6 Astra","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-09-04","last_updated":"2026-09-04","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"openai/gpt-5.6-luna-pro":{"id":"openai/gpt-5.6-luna-pro","name":"GPT 5.6 Luna Pro","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25}},"openai/gpt-4.1-mini":{"id":"openai/gpt-4.1-mini","name":"GPT 4.1 Mini","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"input":1047576,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1}},"openai/gpt-5.4":{"id":"openai/gpt-5.4","name":"GPT 5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":922000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25}},"openai/gpt-oss-20b":{"id":"openai/gpt-oss-20b","name":"GPT OSS 20B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.2,"output":0.3}},"openai/gpt-4-turbo":{"id":"openai/gpt-4-turbo","name":"GPT-4 Turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2023-12","release_date":"2023-11-06","last_updated":"2024-04-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":4096},"cost":{"input":10,"output":30}},"openai/gpt-5.6-sol-pro":{"id":"openai/gpt-5.6-sol-pro","name":"GPT 5.6 Sol Pro","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"openai/gpt-oss-safeguard-20b":{"id":"openai/gpt-oss-safeguard-20b","name":"GPT OSS Safeguard 20B","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"release_date":"2026-02-23","last_updated":"2025-10-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.075,"output":0.3}},"openai/gpt-5.1":{"id":"openai/gpt-5.1","name":"GPT 5.1","description":"Sharper GPT-5 generation for coding, product work, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/gpt-5.1-codex-max":{"id":"openai/gpt-5.1-codex-max","name":"GPT 5.1 Codex Max","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":2.5,"output":20,"cache_read":0.25}},"openai/o1":{"id":"openai/o1","name":"OpenAI o1","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5}},"openai/gpt-4o":{"id":"openai/gpt-4o","name":"GPT-4o","description":"Omni-era GPT for multimodal chat, practical coding, and general assistants","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"openai/gpt-5.6-luna":{"id":"openai/gpt-5.6-luna","name":"GPT 5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25}},"openai/gpt-5.3-codex":{"id":"openai/gpt-5.3-codex","name":"GPT 5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-4o-mini":{"id":"openai/gpt-4o-mini","name":"GPT-4o mini","description":"Small omni GPT for cheap multimodal assistance and production-scale traffic","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"openai/o1-pro":{"id":"openai/o1-pro","name":"OpenAI o1 Pro","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o-pro","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2023-09","release_date":"2025-03-19","last_updated":"2025-03-19","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":150,"output":600,"cache_read":75}},"openai/gpt-4.1":{"id":"openai/gpt-4.1","name":"GPT 4.1","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"input":1047576,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5}},"openai/gpt-5.4-nano":{"id":"openai/gpt-5.4-nano","name":"GPT 5.4 Nano","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02}},"openai/gpt-5.6-terra-pro":{"id":"openai/gpt-5.6-terra-pro","name":"GPT 5.6 Terra Pro","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5}},"openai/gpt-chat-latest":{"id":"openai/gpt-chat-latest","name":"GPT Chat Latest","description":"Chat-tuned GPT model for conversational assistance, writing, and tool workflows","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-03","last_updated":"2026-05-03","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"openai/gpt-5.4-mini":{"id":"openai/gpt-5.4-mini","name":"GPT 5.4 Mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075}},"openai/gpt-3.5-turbo":{"id":"openai/gpt-3.5-turbo","name":"GPT-3.5 Turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2021-09-01","release_date":"2023-03-01","last_updated":"2023-11-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":16385,"input":16385,"output":4096},"cost":{"input":0.5,"output":1.5}},"openai/gpt-5-mini":{"id":"openai/gpt-5-mini","name":"GPT 5 Mini","description":"Small GPT-5 for responsive agents, coding help, and everyday automation","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"openai/gpt-oss-120b":{"id":"openai/gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":0.35,"output":0.75}},"openai/gpt-5.6-terra":{"id":"openai/gpt-5.6-terra","name":"GPT 5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":1050000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5}},"openai/gpt-5.2":{"id":"openai/gpt-5.2","name":"GPT 5.2","description":"Reliable GPT generation for broad coding, writing, and tool-assisted product work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-5":{"id":"openai/gpt-5","name":"GPT 5","description":"Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":400000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/o4-mini":{"id":"openai/o4-mini","name":"OpenAI o4-mini","description":"Fast o-series model for compact reasoning, coding, and tool use","family":"o-mini","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai/o3-mini":{"id":"openai/o3-mini","name":"OpenAI o3-mini","description":"Smaller o-series reasoner for economical coding, math, and planning tasks","family":"o-mini","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai/o3":{"id":"openai/o3","name":"OpenAI o3","description":"Deliberate o-series reasoner for hard math, coding, and multi-step analysis","family":"o","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"input":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":1}},"openai/gpt-5.5":{"id":"openai/gpt-5.5","name":"GPT 5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5}},"openai/gpt-4o-2024-11-20":{"id":"openai/gpt-4o-2024-11-20","name":"GPT-4o (2024-11-20)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-09","release_date":"2024-11-20","last_updated":"2024-11-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"input":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"moonshotai/kimi-latest":{"id":"moonshotai/kimi-latest","name":"Kimi Latest","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-05-03","last_updated":"2026-05-03","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":943718},"cost":{"input":2,"output":10,"cache_read":0.2}},"moonshotai/kimi-k2.7-code-highspeed":{"id":"moonshotai/kimi-k2.7-code-highspeed","name":"Kimi K2.7 Code High-Speed","description":"Lower-latency Kimi Code variant for interactive edits and coding-agent loops","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":1.9,"output":8,"cache_read":0.32}},"moonshotai/kimi-k2.6":{"id":"moonshotai/kimi-k2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":65536},"cost":{"input":0.5,"output":2.6,"cache_read":0.125}},"moonshotai/Kimi-K2-Instruct-0905":{"id":"moonshotai/Kimi-K2-Instruct-0905","name":"Kimi K2 0905","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":100352},"cost":{"input":0.4,"output":1.8,"cache_read":0.2}},"moonshotai/kimi-k2.7-code":{"id":"moonshotai/kimi-k2.7-code","name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":65536},"cost":{"input":0.95,"output":4,"cache_read":0.19}},"moonshotai/kimi-k2-thinking":{"id":"moonshotai/kimi-k2-thinking","name":"Kimi K2 Thinking","description":"Thinking Kimi model for slower research passes, planning, and hard technical questions","family":"kimi-thinking","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-08","release_date":"2025-11-06","last_updated":"2025-11-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":98304},"cost":{"input":0.6,"output":2.5,"cache_read":0.15}},"moonshotai/kimi-k3":{"id":"moonshotai/kimi-k3","name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":943718},"cost":{"input":2,"output":10,"cache_read":0.2}},"moonshotai/kimi-k2-instruct":{"id":"moonshotai/kimi-k2-instruct","name":"Kimi K2 Instruct","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2025-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":8192},"cost":{"input":0.4,"output":1.8,"cache_read":0.2}},"moonshotai/kimi-k2.5":{"id":"moonshotai/kimi-k2.5","name":"Kimi K2.5","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":65536},"cost":{"input":0.3,"output":1.9,"cache_read":0.15}},"moonshotai/kimi-k2.6:thinking":{"id":"moonshotai/kimi-k2.6:thinking","name":"Kimi K2.6 Thinking","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":65536},"cost":{"input":0.5,"output":2.6,"cache_read":0.125}},"moonshotai/kimi-k2-instruct-0711":{"id":"moonshotai/kimi-k2-instruct-0711","name":"Kimi K2 0711","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"release_date":"2024-01-01","last_updated":"2025-07-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.4,"output":1.8,"cache_read":0.2}},"moonshotai/kimi-k2.5:thinking":{"id":"moonshotai/kimi-k2.5:thinking","name":"Kimi K2.5 Thinking","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":65536},"cost":{"input":0.3,"output":1.9,"cache_read":0.15}},"cohere/command-r-plus-08-2024":{"id":"cohere/command-r-plus-08-2024","name":"Cohere: Command R+","description":"Cohere's RAG workhorse for long-context enterprise search and tool use","family":"command-r","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-06-01","release_date":"2024-08-30","last_updated":"2024-08-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":4096},"cost":{"input":2.856,"output":14.246,"cache_read":1.428}},"upstage/solar-pro-3":{"id":"upstage/solar-pro-3","name":"Solar Pro 3","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"solar-pro","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2026-03-03","last_updated":"2026-03-03","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"input":131072,"output":117964},"cost":{"input":0.15,"output":0.6,"cache_read":0.015}},"upstage/solar-pro4":{"id":"upstage/solar-pro4","name":"Solar Pro 4","description":"Upstage's flagship model, specialized for agentic use","family":"solar-pro","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02","release_date":"2026-08-06","last_updated":"2026-08-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":524288,"input":524288,"output":131072},"cost":{"input":0.03,"output":0.12,"cache_read":0.006}},"upstage/solar-pro4:thinking":{"id":"upstage/solar-pro4:thinking","name":"Solar Pro 4 Thinking","description":"Upstage's flagship model, specialized for agentic use","family":"solar-pro","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02","release_date":"2026-08-06","last_updated":"2026-08-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":524288,"input":524288,"output":131072},"cost":{"input":0.03,"output":0.12,"cache_read":0.006}},"arcee-ai/trinity-large-thinking":{"id":"arcee-ai/trinity-large-thinking","name":"Trinity Large Thinking","description":"Reasoning-optimized 398B MoE agent model with extended thinking for long-horizon and multi-turn tool use","family":"trinity","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-01","last_updated":"2026-05-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":80000},"cost":{"input":0.25,"output":0.9,"cache_read":0.125}},"tencent/hy3":{"id":"tencent/hy3","name":"Tencent Hy3","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-06","last_updated":"2026-07-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":262144,"output":128000},"cost":{"input":0.066,"output":0.26,"cache_read":0.029}},"tencent/hy4-preview":{"id":"tencent/hy4-preview","name":"Tencent Hy4 Preview","description":"A next-generation productivity model with significantly enhanced Agent and complex task execution capabilities.","family":"Hy","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-28","last_updated":"2026-08-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":64000},"cost":{"input":0.834,"output":2.501,"cache_read":0.042}},"TheDrummer/skyfall-36b-v2":{"id":"TheDrummer/skyfall-36b-v2","name":"TheDrummer Skyfall 36B V2","description":"Multimodal model for analyzing text, images, documents, and rich media","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-03-10","last_updated":"2025-03-10","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":29491},"cost":{"input":0.55,"output":0.8,"cache_read":0.25}},"TheDrummer/UnslopNemo-12B-v4.1":{"id":"TheDrummer/UnslopNemo-12B-v4.1","name":"UnslopNemo 12b v4","description":"Multimodal model for analyzing text, images, documents, and rich media","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"input":8192,"output":26214},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"TheDrummer/Cydonia-24B-v4.3":{"id":"TheDrummer/Cydonia-24B-v4.3","name":"The Drummer Cydonia 24B v4.3","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-25","last_updated":"2025-12-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.12,"output":0.15,"cache_read":0.06}},"TheDrummer/Cydonia-24B-v2":{"id":"TheDrummer/Cydonia-24B-v2","name":"The Drummer Cydonia 24B v2","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-02-17","last_updated":"2025-02-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.1003,"output":0.1207,"cache_read":0.05015}},"TheDrummer/Cydonia-24B-v4":{"id":"TheDrummer/Cydonia-24B-v4","name":"The Drummer Cydonia 24B v4","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-07-22","last_updated":"2025-07-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.2006,"output":0.2414,"cache_read":0.1003}},"TheDrummer/Anubis-70B-v1.1":{"id":"TheDrummer/Anubis-70B-v1.1","name":"Anubis 70B v1.1","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":16384},"cost":{"input":0.31,"output":0.31,"cache_read":0.155}},"TheDrummer/Magidonia-24B-v4.3":{"id":"TheDrummer/Magidonia-24B-v4.3","name":"The Drummer Magidonia 24B v4.3","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-25","last_updated":"2025-12-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.1003,"output":0.1207,"cache_read":0.05015}},"TheDrummer/Cydonia-24B-v4.1":{"id":"TheDrummer/Cydonia-24B-v4.1","name":"The Drummer Cydonia 24B v4.1","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-08-19","last_updated":"2025-08-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"input":131072,"output":117964},"cost":{"input":0.35,"output":0.55,"cache_read":0.16}},"TheDrummer/Anubis-70B-v1":{"id":"TheDrummer/Anubis-70B-v1","name":"Anubis 70B v1","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":16384},"cost":{"input":0.31,"output":0.31,"cache_read":0.155}},"TheDrummer/Rocinante-12B-v1.1":{"id":"TheDrummer/Rocinante-12B-v1.1","name":"Rocinante 12b","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2024-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"input":16384,"output":8192},"cost":{"input":0.408,"output":0.595,"cache_read":0.204}},"soob3123/Veiled-Calla-12B":{"id":"soob3123/Veiled-Calla-12B","name":"Veiled Calla 12B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-04-13","last_updated":"2025-04-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":8192},"cost":{"input":0.3,"output":0.3,"cache_read":0.15}},"soob3123/amoral-gemma3-27B-v2":{"id":"soob3123/amoral-gemma3-27B-v2","name":"Amoral Gemma3 27B v2","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-05-23","last_updated":"2025-05-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":8192},"cost":{"input":0.3,"output":0.3,"cache_read":0.15}},"soob3123/GrayLine-Qwen3-8B":{"id":"soob3123/GrayLine-Qwen3-8B","name":"Grayline Qwen3 8B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-09-25","last_updated":"2025-09-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.3,"output":0.3,"cache_read":0.15}},"mistral/mistral-medium-3.5:thinking":{"id":"mistral/mistral-medium-3.5:thinking","name":"Mistral Medium 3.5 Thinking","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-04-30","last_updated":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":1.5,"output":7.5,"cache_read":0.75}},"mistral/mistral-medium-3.5":{"id":"mistral/mistral-medium-3.5","name":"Mistral Medium 3.5","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-04-29","last_updated":"2026-04-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":32768},"cost":{"input":1.5,"output":7.5,"cache_read":0.75}},"nanogpt/coding-router:low":{"id":"nanogpt/coding-router:low","name":"Coding Router Low","description":"Automatic model router for matching prompts to suitable backends and budgets","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-05-12","last_updated":"2026-05-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":0.14,"output":0.28,"cache_read":0.028}},"nanogpt/coding-router:high":{"id":"nanogpt/coding-router:high","name":"Coding Router High","description":"Automatic model router for matching prompts to suitable backends and budgets","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-05-12","last_updated":"2026-05-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":1.1,"output":2.2,"cache_read":0.11}},"nanogpt/coding-router":{"id":"nanogpt/coding-router","name":"Coding Router","description":"Automatic model router for matching prompts to suitable backends and budgets","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-05-12","last_updated":"2026-05-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":1.1,"output":2.2,"cache_read":0.11}},"nanogpt/coding-router:max":{"id":"nanogpt/coding-router:max","name":"Coding Router Max","description":"Automatic model router for matching prompts to suitable backends and budgets","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-05-12","last_updated":"2026-05-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5}},"nanogpt/coding-router:medium":{"id":"nanogpt/coding-router:medium","name":"Coding Router Medium","description":"Automatic model router for matching prompts to suitable backends and budgets","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-05-12","last_updated":"2026-05-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":128000},"cost":{"input":0.14,"output":0.28,"cache_read":0.028}},"liquid/lfm-2.5-2.6b":{"id":"liquid/lfm-2.5-2.6b","name":"LFM2.5 2.6B","description":"Liquid AI's compact 2.6B reasoning model for agent workflows, data extraction, RAG, and long-context processing. It supports tool calling and structured output, but Liquid advises against using it for agentic coding. Warning: prompts and responses may be logged and used for model training or service improvement; do not send sensitive data.","family":"liquid","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":32768},"cost":{"input":0.1,"output":0.2,"cache_read":0.05}},"z-ai/glm-4.7":{"id":"z-ai/glm-4.7","name":"GLM 4.7","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":65535},"cost":{"input":0.2,"output":0.8,"cache_read":0.1}},"z-ai/glm-4.5v:thinking":{"id":"z-ai/glm-4.5v:thinking","name":"GLM 4.5V Thinking","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-08-11","last_updated":"2025-08-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":16384},"cost":{"input":0.6,"output":1.8,"cache_read":0.3}},"z-ai/glm-4.6":{"id":"z-ai/glm-4.6","name":"GLM 4.6","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":65535},"cost":{"input":0.35,"output":1.4,"cache_read":0.175}},"z-ai/glm-4.6v":{"id":"z-ai/glm-4.6v","name":"GLM 4.6V","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-08","last_updated":"2025-12-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":24000},"cost":{"input":0.3,"output":0.9,"cache_read":0.15}},"z-ai/glm-4.6v-original":{"id":"z-ai/glm-4.6v-original","name":"GLM 4.6V Original","description":"GLM-4.6V scales its context window to 128k tokens in training, and achieves SoTA performance in visual understanding among models of similar parameter scales. Integrates native Function Calling capabilities, bridging 'visual perception' and 'executable action' for multimodal agents. Direct via Z-AI (Zhipu).","family":"glm","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-12-08","last_updated":"2025-12-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":24000},"cost":{"input":0.6,"output":0.9,"cache_read":0.3}},"z-ai/glm-5.3:thinking":{"id":"z-ai/glm-5.3:thinking","name":"GLM 5.3 Thinking","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"z-ai/GLM-4.6-turbo":{"id":"z-ai/GLM-4.6-turbo","name":"GLM 4.6 Turbo","description":"Fast variant of GLM 4.6 for general chat, coding, and analysis with improved latency and strong reasoning.","family":"glm","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-10-02","last_updated":"2025-10-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"input":204800,"output":131072},"cost":{"input":1,"output":3,"cache_read":0.5}},"z-ai/GLM-4.5-Air:thinking":{"id":"z-ai/GLM-4.5-Air:thinking","name":"GLM 4.5 Air (Thinking)","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":98304},"cost":{"input":0.12,"output":0.8,"cache_read":0.06}},"z-ai/glm-4.6:thinking":{"id":"z-ai/glm-4.6:thinking","name":"GLM 4.6 Thinking","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":65535},"cost":{"input":0.35,"output":1.4,"cache_read":0.175}},"z-ai/glm-5.2":{"id":"z-ai/glm-5.2","name":"GLM 5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.42,"output":1.32,"cache_read":0.078}},"z-ai/glm-4.7-original":{"id":"z-ai/glm-4.7-original","name":"GLM 4.7 Original","description":"GLM-4.7 is a next-gen GLM series text model with stronger reasoning, long-context chat, and reliable tool use. Routed directly via Z-AI (Zhipu).","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":65535},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"z-ai/glm-4.7-flash:thinking":{"id":"z-ai/glm-4.7-flash:thinking","name":"GLM 4.7 Flash Thinking","description":"Budget GLM lane for fast coding help, routing, and everyday automation","family":"glm-flash","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":0.07,"output":0.4,"cache_read":0.035}},"z-ai/glm-4.7:thinking":{"id":"z-ai/glm-4.7:thinking","name":"GLM 4.7 Thinking","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":65535},"cost":{"input":0.2,"output":0.8,"cache_read":0.1}},"z-ai/glm-5v-turbo:thinking":{"id":"z-ai/glm-5v-turbo:thinking","name":"GLM 5V Turbo Thinking","description":"Fast GLM vision model for screenshots, documents, and multimodal agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-01","last_updated":"2026-04-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":202800,"input":202800,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24}},"z-ai/glm-4.7-flash-original":{"id":"z-ai/glm-4.7-flash-original","name":"GLM 4.7 Flash Original","description":"GLM-4.7-Flash is a lightweight 30B model optimized for coding and agentic tasks. Balances high performance with efficiency, perfect for local deployment.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-01-19","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":0.07,"output":0.4,"cache_read":0.035}},"z-ai/GLM-4.5:thinking":{"id":"z-ai/GLM-4.5:thinking","name":"GLM 4.5 (Thinking)","description":"Hybrid-reasoning GLM release that made the 4.5 line broadly useful","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.3,"output":1.3,"cache_read":0.15}},"z-ai/glm-5-original:thinking":{"id":"z-ai/glm-5-original:thinking","name":"GLM 5 Original Thinking","description":"GLM-5 original with extended thinking capabilities for complex reasoning.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-02-11","last_updated":"2026-02-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"z-ai/glm-5.3-flash":{"id":"z-ai/glm-5.3-flash","name":"GLM 5.3 Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.075,"output":0.25,"cache_read":0.015}},"z-ai/glm-4.5":{"id":"z-ai/glm-4.5","name":"GLM 4.5","description":"Hybrid-reasoning GLM release that made the 4.5 line broadly useful","family":"glm","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.3,"output":1.3,"cache_read":0.15}},"z-ai/GLM-4.5-Air":{"id":"z-ai/GLM-4.5-Air","name":"GLM 4.5 Air","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":98304},"cost":{"input":0.12,"output":0.8,"cache_read":0.06}},"z-ai/glm-4.7-original:thinking":{"id":"z-ai/glm-4.7-original:thinking","name":"GLM 4.7 Original Thinking","description":"GLM-4.7 original with extended thinking capabilities for complex reasoning.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":65535},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"z-ai/glm-4.5v":{"id":"z-ai/glm-4.5v","name":"GLM 4.5V","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-08-11","last_updated":"2025-08-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"input":65536,"output":16384},"cost":{"input":0.6,"output":1.8,"cache_read":0.3}},"z-ai/glm-5.1:thinking":{"id":"z-ai/glm-5.1:thinking","name":"GLM 5.1 Thinking","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":131072},"cost":{"input":0.75,"output":2.6,"cache_read":0.15}},"z-ai/glm-5":{"id":"z-ai/glm-5","name":"GLM 5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":0.5,"output":2.55,"cache_read":0.13}},"z-ai/GLM-4.6-turbo:thinking":{"id":"z-ai/GLM-4.6-turbo:thinking","name":"GLM 4.6 Turbo (Thinking)","description":"GLM 4.6 Turbo with thinking mode enabled for enhanced reasoning; shows internal reasoning and supports long context.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"release_date":"2025-10-02","last_updated":"2025-10-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"input":204800,"output":131072},"cost":{"input":1,"output":3,"cache_read":0.5}},"z-ai/glm-5.1":{"id":"z-ai/glm-5.1","name":"GLM 5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":131072},"cost":{"input":0.75,"output":2.6,"cache_read":0.15}},"z-ai/glm-5.2:thinking":{"id":"z-ai/glm-5.2:thinking","name":"GLM 5.2 Thinking","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0.42,"output":1.32,"cache_read":0.078}},"z-ai/glm-latest":{"id":"z-ai/glm-latest","name":"GLM Latest","description":"Compatibility alias that routes to the newest thinking GLM model. Currently routes to GLM 5.2 Thinking.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-05-03","last_updated":"2026-05-03","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"z-ai/glm-5-turbo":{"id":"z-ai/glm-5-turbo","name":"GLM 5 Turbo","description":"Faster GLM-5 lane for coding agents that need lower latency","family":"glm","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-03-16","last_updated":"2026-03-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":202800,"input":202800,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24}},"z-ai/glm-4.7-flash-original:thinking":{"id":"z-ai/glm-4.7-flash-original:thinking","name":"GLM 4.7 Flash Original Thinking","description":"GLM-4.7-Flash with extended thinking capabilities for complex reasoning. Lightweight 30B model optimized for coding and agentic tasks.","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2026-01-19","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":0.07,"output":0.4,"cache_read":0.035}},"z-ai/glm-5-original":{"id":"z-ai/glm-5-original","name":"GLM 5 Original","description":"GLM-5 is Zhipu's latest flagship model with advanced reasoning and instruction following. Routed directly via Z-AI (Zhipu).","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"release_date":"2026-02-11","last_updated":"2026-02-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"z-ai/glm-5.3":{"id":"z-ai/glm-5.3","name":"GLM 5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"z-ai/glm-5v-turbo":{"id":"z-ai/glm-5v-turbo","name":"GLM 5V Turbo","description":"Fast GLM vision model for screenshots, documents, and multimodal agent tasks","family":"glm","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-01","last_updated":"2026-04-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":202800,"input":202800,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24}},"z-ai/glm-5.3-flash-uncensored":{"id":"z-ai/glm-5.3-flash-uncensored","name":"GLM 5.3 Flash Uncensored","description":"GLM 5.3 Flash Uncensored is an uncensored fine-tune of the efficient 320B mixture-of-experts reasoning model, built for unrestricted chat, creative writing, coding, agentic work, tool use, and long-context tasks.","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high"]}],"tool_call":true,"structured_output":true,"release_date":"2026-07-29","last_updated":"2026-08-27","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"input":1048576,"output":32768},"cost":{"input":0.35,"output":1.4,"cache_read":0.175}},"z-ai/glm-4.6-original":{"id":"z-ai/glm-4.6-original","name":"GLM 4.6 Original","description":"GLM-4.6, Zhipu's flagship text model with 256K context window and advanced reasoning capabilities. Direct via Z-AI (Zhipu).","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"input":256000,"output":65535},"cost":{"input":0.35,"output":1.4,"cache_read":0.175}},"z-ai/glm-5:thinking":{"id":"z-ai/glm-5:thinking","name":"GLM 5 Thinking","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":0.5,"output":2.55,"cache_read":0.13}},"z-ai/glm-4.7-flash":{"id":"z-ai/glm-4.7-flash","name":"GLM 4.7 Flash","description":"Budget GLM lane for fast coding help, routing, and everyday automation","family":"glm-flash","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"input":200000,"output":128000},"cost":{"input":0.07,"output":0.4,"cache_read":0.035}},"Doctor-Shotgun/MS3.2-24B-Magnum-Diamond":{"id":"Doctor-Shotgun/MS3.2-24B-Magnum-Diamond","name":"MS3.2 24B Magnum Diamond","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2025-11-24","last_updated":"2025-11-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"input":32768,"output":32768},"cost":{"input":0.493,"output":0.493,"cache_read":0.2465}},"THUDM/GLM-4-9B-0414":{"id":"THUDM/GLM-4-9B-0414","name":"GLM 4 9B 0414","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-04-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32000,"input":32000,"output":8000},"cost":{"input":0.2,"output":0.2,"cache_read":0.1}},"THUDM/GLM-Z1-9B-0414":{"id":"THUDM/GLM-Z1-9B-0414","name":"GLM Z1 9B 0414","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm-z","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-04-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32000,"input":32000,"output":8000},"cost":{"input":0.2,"output":0.2,"cache_read":0.1}},"THUDM/GLM-4-32B-0414":{"id":"THUDM/GLM-4-32B-0414","name":"GLM 4 32B 0414","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"release_date":"2024-01-01","last_updated":"2025-04-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"input":128000,"output":65536},"cost":{"input":0.2,"output":0.2,"cache_read":0.1}}}},"watsonx":{"id":"watsonx","env":["WATSONX_AI_APIKEY","WATSONX_AI_PROJECT_ID"],"npm":"watsonx-ai-provider","name":"watsonx.ai","doc":"https://www.ibm.com/docs/en/watsonx/saas?topic=solutions-supported-foundation-models","models":{"mistralai/mistral-small-3-1-24b-instruct-2503":{"id":"mistralai/mistral-small-3-1-24b-instruct-2503","name":"Mistral Small 3.1 24B","description":"Efficient multimodal model for instruction following, coding, reasoning, and function calling","family":"mistral-small","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-06","release_date":"2025-03-17","last_updated":"2025-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":16384},"cost":{"input":0.106,"output":0.318}},"ibm/granite-4-h-small":{"id":"ibm/granite-4-h-small","name":"Granite-4.0-H-Small","description":"Open-weight hybrid model for enterprise chat, coding, retrieval-augmented generation, and tool-calling workloads","family":"granite","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-02","last_updated":"2025-10-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.0636,"output":0.265}},"meta-llama/llama-4-maverick-17b-128e-instruct-fp8":{"id":"meta-llama/llama-4-maverick-17b-128e-instruct-fp8","name":"Llama 4 Maverick 17B 128E Instruct FP8","description":"Open multimodal Llama for strong reasoning with efficient everyday serving","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":8192},"cost":{"input":0.371,"output":1.484}},"meta-llama/llama-3-3-70b-instruct":{"id":"meta-llama/llama-3-3-70b-instruct","name":"Llama-3.3-70B-Instruct","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":4096},"cost":{"input":0.7526,"output":0.7526}},"openai/gpt-oss-120b":{"id":"openai/gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.159,"output":0.636}}}},"digitalocean":{"id":"digitalocean","env":["DIGITALOCEAN_ACCESS_TOKEN"],"npm":"@ai-sdk/openai-compatible","api":"https://inference.do-ai.run/v1","name":"DigitalOcean","doc":"https://docs.digitalocean.com/products/gradient-ai-platform/details/models/","models":{"openai-gpt-4o":{"id":"openai-gpt-4o","name":"OpenAI GPT-4o","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"openai-gpt-5.2-pro":{"id":"openai-gpt-5.2-pro","name":"OpenAI GPT-5.2 Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":21,"output":168}},"bge-reranker-v2-m3":{"id":"bge-reranker-v2-m3","name":"BGE Reranker v2 M3","description":"Reranking model for improving retrieval quality in search and recommendation systems","family":"bge","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-03-12","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"output":1},"cost":{"input":0.01,"output":0}},"anthropic-claude-opus-4.6":{"id":"anthropic-claude-opus-4.6","name":"Anthropic Claude Opus 4.6","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}}},"openai-o3":{"id":"openai-o3","name":"OpenAI o3","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":0.5}},"deepseek-v4-pro-0813":{"id":"deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":1.32,"output":3.96,"cache_read":0.044}},"deepseek-v4-flash-0731":{"id":"deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":0.08,"output":0.252,"cache_read":0.0252}},"deepseek-v3":{"id":"deepseek-v3","name":"DeepSeek V3","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-07","release_date":"2024-12-26","last_updated":"2025-03-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"output":131072}},"qwen-2.5-14b-instruct":{"id":"qwen-2.5-14b-instruct","name":"Qwen 2.5 14B Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-09","release_date":"2024-09-19","last_updated":"2024-09-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072}},"nvidia-nemotron-3-super-120b":{"id":"nvidia-nemotron-3-super-120b","name":"NVIDIA Nemotron 3 Super 120B (Public Preview)","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-03-11","last_updated":"2026-03-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":32768},"cost":{"input":0.3,"output":0.65,"cache_read":0.06}},"qwen3-coder-flash":{"id":"qwen3-coder-flash","name":"Qwen3 Coder Flash","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.45,"output":1.7,"cache_read":0.09}},"openai-gpt-5.6-terra":{"id":"openai-gpt-5.6-terra","name":"OpenAI GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"gemma-4-31B-it":{"id":"gemma-4-31B-it","name":"Gemma 4","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-22","last_updated":"2026-04-30","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":8192},"cost":{"input":0.18,"output":0.5,"cache_read":0.036}},"alibaba-qwen3-32b":{"id":"alibaba-qwen3-32b","name":"Qwen3 32B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-04-30","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"output":32768},"cost":{"input":0.25,"output":0.55}},"openai-gpt-image-1.5":{"id":"openai-gpt-image-1.5","name":"OpenAI GPT Image 1.5","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","attachment":true,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-11-25","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["image","text"]},"open_weights":false,"limit":{"context":0,"output":16384},"cost":{"input":5,"output":10,"cache_read":1}},"kimi-k2.6":{"id":"kimi-k2.6","name":"Kimi K2.6","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19}},"anthropic-claude-opus-4":{"id":"anthropic-claude-opus-4","name":"Claude Opus 4","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.7,"output":2.2,"cache_read":0.105}},"arcee-trinity-large-thinking":{"id":"arcee-trinity-large-thinking","name":"Arcee Trinity Large Thinking (Public Preview)","description":"Reasoning-optimized 398B MoE agent model with extended thinking for long-horizon and multi-turn tool use","family":"trinity","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"release_date":"2026-04-01","last_updated":"2026-05-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32000},"cost":{"input":0.25,"output":0.9,"cache_read":0.06}},"anthropic-claude-opus-4.7":{"id":"anthropic-claude-opus-4.7","name":"Anthropic Claude Opus 4.7","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic-claude-fable-5":{"id":"anthropic-claude-fable-5","name":"Anthropic Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"release_date":"2026-06-09","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"anthropic-claude-3.5-sonnet":{"id":"anthropic-claude-3.5-sonnet","name":"Claude 3.5 Sonnet","description":"Legacy model retained for compatibility with older integrations","family":"claude-sonnet","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-04","release_date":"2024-06-20","last_updated":"2024-10-22","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"status":"deprecated","cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"minimax-m2.5":{"id":"minimax-m2.5","name":"MiniMax M2.5 (Public Preview)","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax-m2.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-08","release_date":"2026-02-12","last_updated":"2026-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"output":65536},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"openai-gpt-4.1":{"id":"openai-gpt-4.1","name":"OpenAI GPT-4.1","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5}},"openai-gpt-5.3-codex":{"id":"openai-gpt-5.3-codex","name":"OpenAI GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai-gpt-oss-120b":{"id":"openai-gpt-oss-120b","name":"OpenAI GPT-oss-120b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0.055,"output":0.385,"cache_read":0.02}},"openai-gpt-5.2":{"id":"openai-gpt-5.2","name":"OpenAI GPT-5.2","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"anthropic-claude-4.5-haiku":{"id":"anthropic-claude-4.5-haiku","name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":1,"cache_write":1.25}},"e5-large-v2":{"id":"e5-large-v2","name":"E5 Large v2","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2023-05-19","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":512,"output":1024},"cost":{"input":0.02,"output":0}},"openai-gpt-5.6-luna":{"id":"openai-gpt-5.6-luna","name":"OpenAI GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04}}},"anthropic-claude-5-sonnet":{"id":"anthropic-claude-5-sonnet","name":"Anthropic Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"openai-gpt-oss-20b":{"id":"openai-gpt-oss-20b","name":"OpenAI GPT-oss-20b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0.05,"output":0.45}},"deepseek-3.2":{"id":"deepseek-3.2","name":"Deepseek 3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2025-12-02","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"output":163840},"cost":{"input":0.25,"output":0.8,"cache_read":0.075}},"multi-qa-mpnet-base-dot-v1":{"id":"multi-qa-mpnet-base-dot-v1","name":"Multi-QA-mpnet-base-dot-v1","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2021-08-30","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":512,"output":768},"cost":{"input":0.009,"output":0}},"anthropic-claude-3.7-sonnet":{"id":"anthropic-claude-3.7-sonnet","name":"Claude 3.7 Sonnet","description":"Legacy model retained for compatibility with older integrations","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2024-11","release_date":"2025-02-24","last_updated":"2025-02-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"status":"deprecated","cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"nemotron-3-nano-30b":{"id":"nemotron-3-nano-30b","name":"Nemotron 3 Nano 30B A3B","description":"Small Nemotron 3 MoE for efficient coding, math, and long-context agents","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"temperature":true,"release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144}},"gte-large-en-v1.5":{"id":"gte-large-en-v1.5","name":"GTE Large (v1.5)","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-03-27","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"output":1024},"cost":{"input":0.09,"output":0}},"qwen3.5-397b-a17b":{"id":"qwen3.5-397b-a17b","name":"Qwen 3.5 397B A17B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-04-30","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.55,"output":3.5,"cache_read":0.111}},"openai-gpt-5":{"id":"openai-gpt-5","name":"OpenAI GPT-5","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"llama-4-maverick":{"id":"llama-4-maverick","name":"Llama 4 Maverick","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":16384},"cost":{"input":0.2,"output":0.696}},"llama3.3-70b-instruct":{"id":"llama3.3-70b-instruct","name":"Llama 3.3 Instruct (70B)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0.65,"output":0.65}},"all-mini-lm-l6-v2":{"id":"all-mini-lm-l6-v2","name":"All-MiniLM-L6-v2","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2021-08-30","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256,"output":384},"cost":{"input":0.009,"output":0}},"anthropic-claude-sonnet-4":{"id":"anthropic-claude-sonnet-4","name":"Claude Sonnet 4","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.3,"cache_write":3.75,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.3,"cache_write":3.75}}},"kimi-k3":{"id":"kimi-k3","name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":2.85,"output":14.25,"cache_read":0.285}},"openai-gpt-5.4-mini":{"id":"openai-gpt-5.4-mini","name":"OpenAI GPT-5.4 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075}},"anthropic-claude-opus-4.8":{"id":"anthropic-claude-opus-4.8","name":"Anthropic Claude Opus 4.8","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"release_date":"2026-05-28","last_updated":"2026-05-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"openai-gpt-5.4-pro":{"id":"openai-gpt-5.4-pro","name":"OpenAI GPT-5.4 Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}}},"deepseek-4-flash":{"id":"deepseek-4-flash","name":"Deepseek V4 Flash","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"release_date":"2026-05-27","last_updated":"2026-05-29","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":384000},"cost":{"input":0.0679,"output":0.168,"cache_read":0.0168}},"openai-gpt-5.6-sol":{"id":"openai-gpt-5.6-sol","name":"OpenAI GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"tiers":[{"input":8,"output":30,"cache_read":0.8,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8}}},"mistral-nemo-instruct-2407":{"id":"mistral-nemo-instruct-2407","name":"Mistral Nemo Instruct","description":"Legacy model retained for compatibility with older integrations","family":"mistral","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"release_date":"2024-07-18","last_updated":"2024-07-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":16384},"status":"deprecated","cost":{"input":0.3,"output":0.3}},"nemotron-3-ultra-550b":{"id":"nemotron-3-ultra-550b","name":"Nemotron 3 Ultra","description":"Flagship Nemotron model for high-throughput reasoning and complex agents","family":"nemotron","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"release_date":"2026-06-04","last_updated":"2026-06-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":131072},"cost":{"input":0.9,"output":1.7}},"anthropic-claude-fable-5.1":{"id":"anthropic-claude-fable-5.1","name":"Anthropic Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5}},"nemotron-nano-12b-v2-vl":{"id":"nemotron-nano-12b-v2-vl","name":"Nemotron-nano 12b v2-vl","description":"Nemotron multimodal model for visual reasoning and agentic AI workflows","family":"nemotron","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"temperature":true,"release_date":"2025-10-28","last_updated":"2025-10-28","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":16384},"cost":{"input":0.2,"output":0.6}},"glm-5.3-flash":{"id":"glm-5.3-flash","name":"GLM5.3 Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":1048576},"cost":{"input":0.15,"output":0.5,"cache_read":0.03}},"deepseek-r1-distill-llama-70b":{"id":"deepseek-r1-distill-llama-70b","name":"DeepSeek R1 Distill Llama 70B","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-01-30","last_updated":"2025-01-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32678,"output":8192},"cost":{"input":0.99,"output":0.99}},"openai-o3-mini":{"id":"openai-o3-mini","name":"OpenAI o3 mini","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai-gpt-5.1-codex-max":{"id":"openai-gpt-5.1-codex-max","name":"GPT-5.1 Codex Max","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"nemotron-3-nano-omni":{"id":"nemotron-3-nano-omni","name":"Nemotron 3 Nano Omni","description":"Open Nemotron omni model combining reasoning with text, vision, and audio","family":"nemotron","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-28","last_updated":"2026-04-28","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"output":65536},"cost":{"input":0.5,"output":0.9}},"openai-gpt-5.4":{"id":"openai-gpt-5.4","name":"OpenAI GPT-5.4","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}}},"openai-gpt-5-nano":{"id":"openai-gpt-5-nano","name":"OpenAI GPT-5 Nano","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005}},"anthropic-claude-4.5-sonnet":{"id":"anthropic-claude-4.5-sonnet","name":"Anthropic Claude 4.5 Sonnet","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}}},"mistral-7b-instruct-v0.3":{"id":"mistral-7b-instruct-v0.3","name":"Mistral 7B Instruct v0.3","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"release_date":"2024-05-22","last_updated":"2024-05-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"output":32768}},"ministral-3-8b-instruct-2512":{"id":"ministral-3-8b-instruct-2512","name":"Ministral 3 8B","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"release_date":"2025-12-15","last_updated":"2025-12-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144}},"openai-gpt-5.5":{"id":"openai-gpt-5.5","name":"OpenAI GPT-5.5","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}}},"glm-5":{"id":"glm-5","name":"GLM 5","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"release_date":"2026-02-11","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":64000,"output":64000},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"openai-gpt-5-mini":{"id":"openai-gpt-5-mini","name":"OpenAI GPT-5 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"anthropic-claude-3.5-haiku":{"id":"anthropic-claude-3.5-haiku","name":"Claude 3.5 Haiku","description":"Legacy model retained for compatibility with older integrations","family":"claude-haiku","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-07","release_date":"2024-11-05","last_updated":"2024-11-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"status":"deprecated","cost":{"input":0.8,"output":4,"cache_read":0.08,"cache_write":1}},"qwen3.8-max":{"id":"qwen3.8-max","name":"Qwen3.8-Max","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":262144},"cost":{"input":2,"output":6,"cache_read":0.2}},"kimi-k2.5":{"id":"kimi-k2.5","name":"Kimi K2.5","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":2.7,"cache_read":0.203}},"glm-5.1":{"id":"glm-5.1","name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"output":163840},"cost":{"input":1.3,"output":4.3,"cache_read":0.26}},"mistral-3-14B":{"id":"mistral-3-14B","name":"Ministral 3 14B","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-12-15","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":128000},"cost":{"input":0.2,"output":0.2}},"qwen3-tts-voicedesign":{"id":"qwen3-tts-voicedesign","name":"Qwen3 TTS VoiceDesign","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2026-04-21","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["audio"]},"open_weights":true,"limit":{"context":32768,"output":1}},"bge-m3":{"id":"bge-m3","name":"BGE M3","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"bge","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-01-30","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"output":1024},"cost":{"input":0.02,"output":0}},"qwen3-embedding-0.6b":{"id":"qwen3-embedding-0.6b","name":"Qwen3 Embedding 0.6B","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-06-03","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":8000,"output":1024},"status":"beta","cost":{"input":0.04,"output":0}},"openai-o1":{"id":"openai-o1","name":"OpenAI o1","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5}},"wan2-2-t2v-a14b":{"id":"wan2-2-t2v-a14b","name":"Wan2.2-T2V-A14B","description":"Video model for prompt-guided generation, editing, and motion workflows","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-07-28","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["video"]},"open_weights":true,"limit":{"context":100,"output":1},"cost":{"input":0.6,"output":0}},"anthropic-claude-3-opus":{"id":"anthropic-claude-3-opus","name":"Claude 3 Opus","description":"Legacy model retained for compatibility with older integrations","family":"claude-opus","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-08","release_date":"2024-02-29","last_updated":"2024-02-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":4096},"status":"deprecated","cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75}},"anthropic-claude-4.1-opus":{"id":"anthropic-claude-4.1-opus","name":"Anthropic Claude 4.1 Opus","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75}},"openai-gpt-image-1":{"id":"openai-gpt-image-1","name":"GPT Image 1","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","attachment":true,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-04-24","last_updated":"2025-04-24","modalities":{"input":["text","image"],"output":["image"]},"open_weights":false,"limit":{"context":0,"output":0},"cost":{"input":5,"output":40,"cache_read":1.25}},"deepseek-v4-pro":{"id":"deepseek-v4-pro","name":"Deepseek V4 Pro","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":384000},"cost":{"input":0.87,"output":1.74,"cache_read":0.174}},"stable-diffusion-3.5-large":{"id":"stable-diffusion-3.5-large","name":"Stable Diffusion 3.5 Large","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"stable-diffusion","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-10-22","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["image"]},"open_weights":true,"limit":{"context":256,"output":1},"cost":{"input":0.08,"output":0}},"anthropic-claude-opus-4.5":{"id":"anthropic-claude-opus-4.5","name":"Anthropic Claude Opus 4.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-11-24","last_updated":"2025-11-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"llama3-8b-instruct":{"id":"llama3-8b-instruct","name":"Llama 3.1 Instruct (8B)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.198,"output":0.198}},"glm-5.3":{"id":"glm-5.3","name":"GLM5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"anthropic-claude-opus-5":{"id":"anthropic-claude-opus-5","name":"Anthropic Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"openai-gpt-5.4-nano":{"id":"openai-gpt-5.4-nano","name":"OpenAI GPT-5.4 Nano","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02}},"anthropic-claude-4.6-sonnet":{"id":"anthropic-claude-4.6-sonnet","name":"Anthropic Claude Sonnet 4.6","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}}},"anthropic-claude-haiku-4.5":{"id":"anthropic-claude-haiku-4.5","name":"Anthropic Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"mimo-v2.5-pro":{"id":"mimo-v2.5-pro","name":"MiMo V2.5 Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":1.5,"cache_read":0.08}},"openai-gpt-image-2":{"id":"openai-gpt-image-2","name":"OpenAI GPT Image 2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","attachment":true,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-04-24","last_updated":"2025-04-24","modalities":{"input":["text","image"],"output":["image","text"]},"open_weights":false,"limit":{"context":0,"output":16384},"cost":{"input":8,"output":30}},"openai-gpt-4o-mini":{"id":"openai-gpt-4o-mini","name":"OpenAI GPT-4o mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"fal-ai/fast-sdxl":{"id":"fal-ai/fast-sdxl","name":"Fast SDXL","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"stable-diffusion","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2023-07-26","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["image"]},"open_weights":true,"limit":{"context":0,"output":0}},"fal-ai/elevenlabs/tts/multilingual-v2":{"id":"fal-ai/elevenlabs/tts/multilingual-v2","name":"ElevenLabs Multilingual TTS v2","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"elevenlabs","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2023-08-22","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["audio"]},"open_weights":false,"limit":{"context":0,"output":0}},"fal-ai/stable-audio-25/text-to-audio":{"id":"fal-ai/stable-audio-25/text-to-audio","name":"Stable Audio 2.5 (Text-to-Audio)","description":"Speech generation model for controllable voice, narration, and audio delivery","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-10-08","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["audio"]},"open_weights":false,"limit":{"context":0,"output":0}},"fal-ai/flux/schnell":{"id":"fal-ai/flux/schnell","name":"FLUX.1 [schnell]","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"flux","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-08-01","last_updated":"2026-04-16","modalities":{"input":["text"],"output":["image"]},"open_weights":true,"limit":{"context":0,"output":0}}}},"vivgrid":{"id":"vivgrid","env":["VIVGRID_API_KEY"],"npm":"@ai-sdk/openai","api":"https://api.vivgrid.com/v1","name":"Vivgrid","doc":"https://docs.vivgrid.com/models","models":{"gpt-5.1-codex":{"id":"gpt-5.1-codex","name":"GPT-5.1 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"gpt-5.6-sol":{"id":"gpt-5.6-sol","name":"GPT 5.6 Sol","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"cache_write":6.25}},"gemini-3.1-pro-preview":{"id":"gemini-3.1-pro-preview","name":"Gemini 3.1 Pro Preview","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"gpt-5.2-codex":{"id":"gpt-5.2-codex","name":"GPT-5.2 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-01-14","last_updated":"2026-01-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":128000},"cost":{"input":1.2,"output":4.2,"cache_read":0.3}},"deepseek-v4-flash":{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.15,"output":0.3,"reasoning":0.3,"cache_read":0.03}},"gpt-5.4":{"id":"gpt-5.4","name":"GPT-5.4","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":2.5,"output":15,"cache_read":0.25}},"gpt-5.1-codex-max":{"id":"gpt-5.1-codex-max","name":"GPT-5.1 Codex Max","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"gemini-3.1-flash-lite-preview":{"id":"gemini-3.1-flash-lite-preview","name":"Gemini 3.1 Flash Lite Preview","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-03-03","last_updated":"2026-03-03","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"cache_write":1}},"gpt-5.6-luna":{"id":"gpt-5.6-luna","name":"GPT 5.6 Luna","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":6,"cache_read":0.1,"cache_write":1.25}},"kimi-k3":{"id":"kimi-k3","name":"Kimi K3","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3}},"deepseek-v3.2":{"id":"deepseek-v3.2","name":"DeepSeek-V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":0.28,"output":0.42}},"gpt-5.3-codex":{"id":"gpt-5.3-codex","name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-02-24","last_updated":"2026-02-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"glm-5.3-flash":{"id":"glm-5.3-flash","name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.04}},"gpt-5.4-nano":{"id":"gpt-5.4-nano","name":"GPT-5.4 Nano","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":0.2,"output":1.25,"cache_read":0.02}},"gpt-5.4-mini":{"id":"gpt-5.4-mini","name":"GPT-5.4 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":0.75,"output":4.5,"cache_read":0.075}},"deepseek-v4-pro":{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":0.435,"output":0.87,"cache_read":0.003625}},"gpt-5-mini":{"id":"gpt-5-mini","name":"GPT-5 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":272000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":0.25,"output":2,"cache_read":0.03}},"gemini-3.7-flash":{"id":"gemini-3.7-flash","name":"Gemini 3.7 Flash","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":0.75,"output":3.75,"cache_read":0.075}},"gpt-5.6-terra":{"id":"gpt-5.6-terra","name":"GPT 5.6 Terra","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"cache_write":3.125}},"glm-5.3":{"id":"glm-5.3","name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.2,"output":4.2,"cache_read":0.26}},"gpt-5.5":{"id":"gpt-5.5","name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible"},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}}}}},"auriko":{"id":"auriko","env":["AURIKO_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.auriko.ai/v1","name":"Auriko","doc":"https://docs.auriko.ai","models":{"claude-sonnet-4-6":{"id":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"minimax-m2-7-highspeed":{"id":"minimax-m2-7-highspeed","name":"MiniMax-M2.7-highspeed","description":"Low-latency M2.7 variant for interactive coding plans and agent loops","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_write":0.375}},"grok-4.3":{"id":"grok-4.3","name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"kimi-k2.6":{"id":"kimi-k2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.16}},"gemini-3.1-pro-preview":{"id":"gemini-3.1-pro-preview","name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"qwen-3.6-plus":{"id":"qwen-3.6-plus","name":"Qwen3.6 Plus","description":"Earlier Qwen multimodal workhorse for million-token agent and document tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.1,"tiers":[{"input":2,"output":6,"cache_read":0.2,"cache_write":2.5,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2,"output":6,"cache_read":0.2,"cache_write":2.5}}},"deepseek-v4-flash":{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028}},"claude-opus-4-6":{"id":"claude-opus-4-6","name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"claude-opus-4-7":{"id":"claude-opus-4-7","name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"experimental":{"modes":{"fast":{"cost":{"input":30,"output":150,"cache_read":3,"cache_write":37.5},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"kimi-k2.5":{"id":"kimi-k2.5","name":"Kimi K2.5","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":2.8}},"glm-5.1":{"id":"glm-5.1","name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"deepseek-v4-pro":{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.435,"output":0.87,"cache_read":0.003625}},"gemini-2.5-pro":{"id":"gemini-2.5-pro","name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}}},"minimax-m2-7":{"id":"minimax-m2-7","name":"MiniMax-M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_write":0.375}},"gemini-2.5-flash":{"id":"gemini-2.5-flash","name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}}}},"siliconflow-cn":{"id":"siliconflow-cn","env":["SILICONFLOW_CN_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.siliconflow.cn/v1","name":"SiliconFlow (China)","doc":"https://cloud.siliconflow.com/models","models":{"baidu/ERNIE-4.5-300B-A47B":{"id":"baidu/ERNIE-4.5-300B-A47B","name":"baidu/ERNIE-4.5-300B-A47B","description":"Tool-capable chat model for instruction following and agentic application workflows","family":"ernie","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-02","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":131000},"cost":{"input":0.28,"output":1.1}},"stepfun-ai/Step-3.5-Flash":{"id":"stepfun-ai/Step-3.5-Flash","name":"stepfun-ai/Step-3.5-Flash","description":"StepFun flash model for efficient multimodal reasoning, coding, and tool use","family":"step","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-11","last_updated":"2026-02-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.1,"output":0.3}},"deepseek-ai/DeepSeek-V4-Flash":{"id":"deepseek-ai/DeepSeek-V4-Flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.003}},"deepseek-ai/DeepSeek-V3.1-Terminus":{"id":"deepseek-ai/DeepSeek-V3.1-Terminus","name":"deepseek-ai/DeepSeek-V3.1-Terminus","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-09-29","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.27,"output":1}},"deepseek-ai/DeepSeek-OCR":{"id":"deepseek-ai/DeepSeek-OCR","name":"deepseek-ai/DeepSeek-OCR","description":"OCR model for extracting structured text from documents and screenshots","attachment":true,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2025-10-20","last_updated":"2025-10-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"output":8192},"cost":{"input":0,"output":0}},"deepseek-ai/DeepSeek-R1":{"id":"deepseek-ai/DeepSeek-R1","name":"deepseek-ai/DeepSeek-R1","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-05-28","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.5,"output":2.18}},"deepseek-ai/DeepSeek-V3.2":{"id":"deepseek-ai/DeepSeek-V3.2","name":"deepseek-ai/DeepSeek-V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-12-03","last_updated":"2025-12-03","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.27,"output":0.42}},"deepseek-ai/DeepSeek-V4-Pro":{"id":"deepseek-ai/DeepSeek-V4-Pro","name":"deepseek-ai/DeepSeek-V4-Pro","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1049000,"output":393000},"cost":{"input":1.74,"output":3.48,"cache_read":0.145}},"deepseek-ai/DeepSeek-V3":{"id":"deepseek-ai/DeepSeek-V3","name":"deepseek-ai/DeepSeek-V3","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2024-12-26","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.25,"output":1}},"inclusionAI/Ling-flash-2.0":{"id":"inclusionAI/Ling-flash-2.0","name":"inclusionAI/Ling-flash-2.0","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-09-18","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":131000},"cost":{"input":0.14,"output":0.57}},"zai-org/GLM-5.2":{"id":"zai-org/GLM-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1049000,"output":262000},"cost":{"input":1.4,"output":4.4,"cache_read":0.26,"cache_write":0}},"zai-org/GLM-4.5-Air":{"id":"zai-org/GLM-4.5-Air","name":"zai-org/GLM-4.5-Air","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm-air","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-28","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":131000},"cost":{"input":0.14,"output":0.86}},"Qwen/Qwen3.5-27B":{"id":"Qwen/Qwen3.5-27B","name":"Qwen/Qwen3.5-27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-02-25","last_updated":"2026-02-25","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.26,"output":2.09}},"Qwen/Qwen3-8B":{"id":"Qwen/Qwen3-8B","name":"Qwen/Qwen3-8B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-04-30","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":131000},"cost":{"input":0.06,"output":0.06}},"Qwen/Qwen3-14B":{"id":"Qwen/Qwen3-14B","name":"Qwen/Qwen3-14B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-04-30","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":131000},"cost":{"input":0.07,"output":0.28}},"Qwen/Qwen3.5-4B":{"id":"Qwen/Qwen3.5-4B","name":"Qwen/Qwen3.5-4B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-03-03","last_updated":"2026-03-03","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0,"output":0}},"Qwen/Qwen3.5-9B":{"id":"Qwen/Qwen3.5-9B","name":"Qwen/Qwen3.5-9B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-03-03","last_updated":"2026-03-03","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.22,"output":1.74}},"Qwen/Qwen3.5-122B-A10B":{"id":"Qwen/Qwen3.5-122B-A10B","name":"Qwen/Qwen3.5-122B-A10B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-02-26","last_updated":"2026-02-26","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.29,"output":2.32}},"Qwen/Qwen3.5-397B-A17B":{"id":"Qwen/Qwen3.5-397B-A17B","name":"Qwen/Qwen3.5-397B-A17B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-02-16","last_updated":"2026-02-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.29,"output":1.74}},"Qwen/Qwen3.5-35B-A3B":{"id":"Qwen/Qwen3.5-35B-A3B","name":"Qwen/Qwen3.5-35B-A3B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-02-25","last_updated":"2026-02-25","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.23,"output":1.86}},"Qwen/Qwen3-32B":{"id":"Qwen/Qwen3-32B","name":"Qwen/Qwen3-32B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-04-30","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":131000},"cost":{"input":0.14,"output":0.57}},"Qwen/Qwen3-235B-A22B-Thinking-2507":{"id":"Qwen/Qwen3-235B-A22B-Thinking-2507","name":"Qwen/Qwen3-235B-A22B-Thinking-2507","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-28","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.13,"output":0.6}},"Qwen/Qwen3.6-35B-A3B":{"id":"Qwen/Qwen3.6-35B-A3B","name":"Qwen/Qwen3.6-35B-A3B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.23,"output":1.86}},"Qwen/Qwen3-VL-30B-A3B-Instruct":{"id":"Qwen/Qwen3-VL-30B-A3B-Instruct","name":"Qwen/Qwen3-VL-30B-A3B-Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-05","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.29,"output":1}},"Qwen/Qwen3-VL-235B-A22B-Thinking":{"id":"Qwen/Qwen3-VL-235B-A22B-Thinking","name":"Qwen/Qwen3-VL-235B-A22B-Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-04","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.45,"output":3.5}},"Qwen/Qwen3-VL-8B-Instruct":{"id":"Qwen/Qwen3-VL-8B-Instruct","name":"Qwen/Qwen3-VL-8B-Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-15","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.18,"output":0.68}},"Qwen/Qwen3-VL-32B-Instruct":{"id":"Qwen/Qwen3-VL-32B-Instruct","name":"Qwen/Qwen3-VL-32B-Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-21","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.2,"output":0.6}},"Qwen/Qwen2.5-72B-Instruct":{"id":"Qwen/Qwen2.5-72B-Instruct","name":"Qwen/Qwen2.5-72B-Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2024-09-18","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":33000,"output":4000},"cost":{"input":0.59,"output":0.59}},"Qwen/Qwen2.5-7B-Instruct":{"id":"Qwen/Qwen2.5-7B-Instruct","name":"Qwen/Qwen2.5-7B-Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2024-09-18","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":33000,"output":4000},"cost":{"input":0.05,"output":0.05}},"Qwen/Qwen3-Coder-480B-A35B-Instruct":{"id":"Qwen/Qwen3-Coder-480B-A35B-Instruct","name":"Qwen/Qwen3-Coder-480B-A35B-Instruct","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-31","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.25,"output":1}},"Qwen/Qwen3-VL-235B-A22B-Instruct":{"id":"Qwen/Qwen3-VL-235B-A22B-Instruct","name":"Qwen/Qwen3-VL-235B-A22B-Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-04","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.3,"output":1.5}},"Qwen/Qwen3-Coder-30B-A3B-Instruct":{"id":"Qwen/Qwen3-Coder-30B-A3B-Instruct","name":"Qwen/Qwen3-Coder-30B-A3B-Instruct","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-01","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.07,"output":0.28}},"Qwen/Qwen3-VL-32B-Thinking":{"id":"Qwen/Qwen3-VL-32B-Thinking","name":"Qwen/Qwen3-VL-32B-Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-21","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.2,"output":1.5}},"Qwen/Qwen3-VL-30B-A3B-Thinking":{"id":"Qwen/Qwen3-VL-30B-A3B-Thinking","name":"Qwen/Qwen3-VL-30B-A3B-Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-11","last_updated":"2025-11-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.29,"output":1}},"Qwen/Qwen3-30B-A3B-Instruct-2507":{"id":"Qwen/Qwen3-30B-A3B-Instruct-2507","name":"Qwen/Qwen3-30B-A3B-Instruct-2507","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-30","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.09,"output":0.3}},"ByteDance-Seed/Seed-OSS-36B-Instruct":{"id":"ByteDance-Seed/Seed-OSS-36B-Instruct","name":"ByteDance-Seed/Seed-OSS-36B-Instruct","description":"Tool-capable chat model for instruction following and agentic application workflows","family":"seed","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-09-04","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262000,"output":262000},"cost":{"input":0.21,"output":0.57}},"Pro/deepseek-ai/DeepSeek-V3":{"id":"Pro/deepseek-ai/DeepSeek-V3","name":"Pro/deepseek-ai/DeepSeek-V3","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2024-12-26","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.25,"output":1}},"Pro/deepseek-ai/DeepSeek-V3.1-Terminus":{"id":"Pro/deepseek-ai/DeepSeek-V3.1-Terminus","name":"Pro/deepseek-ai/DeepSeek-V3.1-Terminus","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-09-29","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.27,"output":1}},"Pro/deepseek-ai/DeepSeek-R1":{"id":"Pro/deepseek-ai/DeepSeek-R1","name":"Pro/deepseek-ai/DeepSeek-R1","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-05-28","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.5,"output":2.18}},"Pro/deepseek-ai/DeepSeek-V3.2":{"id":"Pro/deepseek-ai/DeepSeek-V3.2","name":"Pro/deepseek-ai/DeepSeek-V3.2","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-12-03","last_updated":"2025-12-03","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":164000,"output":164000},"cost":{"input":0.27,"output":0.42}},"Pro/zai-org/GLM-5.1":{"id":"Pro/zai-org/GLM-5.1","name":"Pro/zai-org/GLM-5.1","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-08","last_updated":"2026-04-08","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":205000,"output":205000},"cost":{"input":1.4,"output":4.4,"cache_read":0.26,"cache_write":0}},"Pro/zai-org/GLM-5":{"id":"Pro/zai-org/GLM-5","name":"Pro/zai-org/GLM-5","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":205000,"output":205000},"cost":{"input":1,"output":3.2}},"Pro/MiniMaxAI/MiniMax-M2.5":{"id":"Pro/MiniMaxAI/MiniMax-M2.5","name":"Pro/MiniMaxAI/MiniMax-M2.5","description":"Frontier MiniMax model for engineering, office tasks, and agentic reasoning","family":"minimax","attachment":false,"reasoning":false,"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-13","last_updated":"2026-02-13","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":192000,"output":131000},"cost":{"input":0.3,"output":1.22}},"Pro/moonshotai/Kimi-K2.5":{"id":"Pro/moonshotai/Kimi-K2.5","name":"Pro/moonshotai/Kimi-K2.5","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-01-27","last_updated":"2026-01-27","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262000,"output":262000},"cost":{"input":0.45,"output":2.25,"cache_read":0.07}},"Pro/moonshotai/Kimi-K2.6":{"id":"Pro/moonshotai/Kimi-K2.6","name":"Pro/moonshotai/Kimi-K2.6","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262000,"output":262000},"cost":{"input":0.95,"output":4,"cache_read":0.16}},"tencent/Hunyuan-A13B-Instruct":{"id":"tencent/Hunyuan-A13B-Instruct","name":"tencent/Hunyuan-A13B-Instruct","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"hunyuan","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-06-30","last_updated":"2025-11-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131000,"output":131000},"cost":{"input":0.14,"output":0.57}},"PaddlePaddle/PaddleOCR-VL-1.5":{"id":"PaddlePaddle/PaddleOCR-VL-1.5","name":"PaddlePaddle/PaddleOCR-VL-1.5","description":"Multimodal model for analyzing text, images, documents, and rich media","attachment":true,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-01-29","last_updated":"2026-01-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":16384,"output":16384},"cost":{"input":0,"output":0}}}},"nova":{"id":"nova","env":["NOVA_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.nova.amazon.com/v1","name":"Nova","doc":"https://nova.amazon.com/dev/documentation","models":{"nova-2-lite-v1":{"id":"nova-2-lite-v1","name":"Nova 2 Lite","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":0,"output":0,"reasoning":0}},"nova-2-pro-v1":{"id":"nova-2-pro-v1","name":"Nova 2 Pro","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova-pro","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2025-12-03","last_updated":"2026-01-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":0,"output":0,"reasoning":0}}}},"inceptron":{"id":"inceptron","env":["INCEPTRON_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.inceptron.io/v1","name":"Inceptron","doc":"https://docs.inceptron.io","models":{"deepseek-ai/DeepSeek-V4-Flash-0731":{"id":"deepseek-ai/DeepSeek-V4-Flash-0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":0.13,"output":0.28,"cache_read":0.03,"cache_write":0}},"zai-org/GLM-5.2":{"id":"zai-org/GLM-5.2","name":"GLM 5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":0.71,"output":2.35,"cache_read":0.12,"cache_write":0}},"moonshotai/Kimi-K2.7-Code":{"id":"moonshotai/Kimi-K2.7-Code","name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.66,"output":3.4,"cache_read":0.18,"cache_write":0}},"moonshotai/Kimi-K2.6":{"id":"moonshotai/Kimi-K2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.53,"output":3.39,"cache_read":0.17,"cache_write":0}}}},"vultr":{"id":"vultr","env":["VULTR_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.vultrinference.com/v1","name":"Vultr","doc":"https://api.vultrinference.com/","models":{"deepseek-ai/DeepSeek-V4-Flash":{"id":"deepseek-ai/DeepSeek-V4-Flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.3,"output":1}},"nvidia/DeepSeek-V3.2-NVFP4":{"id":"nvidia/DeepSeek-V3.2-NVFP4","name":"DeepSeek V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.55,"output":1.65}},"nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16":{"id":"nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16","name":"NVIDIA Nemotron 3 Nano Omni","description":"Open Nemotron omni model combining reasoning with text, vision, and audio","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-28","last_updated":"2026-04-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":131072},"cost":{"input":0.13,"output":0.38}},"nvidia/Nemotron-Cascade-2-30B-A3B":{"id":"nvidia/Nemotron-Cascade-2-30B-A3B","name":"NVIDIA Nemotron Cascade 2","description":"Nemotron model for efficient reasoning, coding, and specialized AI agents","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":131072},"cost":{"input":0.15,"output":0.6}},"zai-org/GLM-5.2-FP8":{"id":"zai-org/GLM-5.2-FP8","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":393216,"output":131072},"cost":{"input":0.85,"output":3.1}},"Qwen/Qwen3.5-397B-A17B":{"id":"Qwen/Qwen3.5-397B-A17B","name":"Qwen3.5 397B-A17B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.3,"output":2}},"Qwen/Qwen3.6-27B":{"id":"Qwen/Qwen3.6-27B","name":"Qwen3.6 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.3,"output":2}},"MiniMaxAI/MiniMax-M2.7":{"id":"MiniMaxAI/MiniMax-M2.7","name":"MiniMax-M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2}},"moonshotai/Kimi-K2.6":{"id":"moonshotai/Kimi-K2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":131072},"cost":{"input":0.3,"output":1.2}},"XiaomiMiMo/MiMo-V2.5-Pro":{"id":"XiaomiMiMo/MiMo-V2.5-Pro","name":"MiMo-V2.5-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.55,"output":1.65}}}},"ollama-cloud":{"id":"ollama-cloud","env":["OLLAMA_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://ollama.com/v1","name":"Ollama Cloud","doc":"https://docs.ollama.com/cloud","models":{"gpt-oss:20b":{"id":"gpt-oss:20b","name":"gpt-oss:20b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"release_date":"2025-08-05","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768}},"deepseek-v4-flash:0731":{"id":"deepseek-v4-flash:0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576}},"minimax-m2.7":{"id":"minimax-m2.7","name":"minimax-m2.7","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":196608,"output":196608}},"kimi-k2.6":{"id":"kimi-k2.6","name":"kimi-k2.6","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"release_date":"2026-04-20","last_updated":"2026-04-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":976000,"output":131072}},"minimax-m2.5":{"id":"minimax-m2.5","name":"minimax-m2.5","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"knowledge":"2025-01","release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072}},"minimax-m3":{"id":"minimax-m3","name":"minimax-m3","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax-m3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-31","last_updated":"2026-05-31","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":512000,"output":131072}},"qwen3.5:397b":{"id":"qwen3.5:397b","name":"qwen3.5:397b","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_details"},"release_date":"2026-02-15","last_updated":"2026-02-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536}},"deepseek-v4-flash":{"id":"deepseek-v4-flash","name":"deepseek-v4-flash","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576}},"kimi-k2.7-code":{"id":"kimi-k2.7-code","name":"kimi-k2.7-code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144}},"gpt-oss:120b":{"id":"gpt-oss:120b","name":"gpt-oss:120b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"release_date":"2025-08-05","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768}},"nemotron-3-ultra":{"id":"nemotron-3-ultra","name":"nemotron-3-ultra","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2026-06-04","last_updated":"2026-06-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":128000}},"nemotron-3-nano:30b":{"id":"nemotron-3-nano:30b","name":"nemotron-3-nano:30b","description":"Small Nemotron 3 MoE for efficient coding, math, and long-context agents","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2025-12-15","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072}},"mistral-large-3:675b":{"id":"mistral-large-3:675b","name":"mistral-large-3:675b","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","attachment":true,"reasoning":false,"tool_call":true,"release_date":"2025-12-02","last_updated":"2026-01-19","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144}},"kimi-k3":{"id":"kimi-k3","name":"kimi-k3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-27","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072}},"glm-5.3-flash":{"id":"glm-5.3-flash","name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072}},"gemma4:31b":{"id":"gemma4:31b","name":"gemma4:31b","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144}},"kimi-k2.5":{"id":"kimi-k2.5","name":"kimi-k2.5","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"release_date":"2026-01-27","last_updated":"2026-01-27","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144}},"glm-5.1":{"id":"glm-5.1","name":"glm-5.1","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"release_date":"2026-03-27","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":131072}},"deepseek-v4-pro":{"id":"deepseek-v4-pro","name":"deepseek-v4-pro","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576}},"glm-5.3":{"id":"glm-5.3","name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072}},"nemotron-3-super":{"id":"nemotron-3-super","name":"nemotron-3-super","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2026-03-11","last_updated":"2026-03-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536}}}},"freemodel":{"id":"freemodel","env":["FREEMODEL_API_KEY"],"npm":"@ai-sdk/anthropic","api":"https://cc.freemodel.dev/v1","name":"FreeModel","doc":"https://freemodel.dev","models":{"claude-sonnet-4-6":{"id":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"gpt-5.4":{"id":"gpt-5.4","name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://api.freemodel.dev/v1"},"cost":{"input":2.5,"output":15,"cache_read":0.25,"cache_write":2.5}},"claude-opus-4-6":{"id":"claude-opus-4-6","name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"claude-opus-4-7":{"id":"claude-opus-4-7","name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"gpt-5.3-codex":{"id":"gpt-5.3-codex","name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://api.freemodel.dev/v1"},"cost":{"input":1.75,"output":14,"cache_read":0.175,"cache_write":1.75}},"claude-haiku-4-5-20251001":{"id":"claude-haiku-4-5-20251001","name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"claude-fable-5":{"id":"claude-fable-5","name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"gpt-5.4-mini":{"id":"gpt-5.4-mini","name":"GPT-5.4 mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://api.freemodel.dev/v1"},"cost":{"input":0.75,"output":4.5,"cache_read":0.075,"cache_write":0.75}},"claude-opus-4-8":{"id":"claude-opus-4-8","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"gpt-5.5":{"id":"gpt-5.5","name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://api.freemodel.dev/v1"},"cost":{"input":5,"output":30,"cache_read":0.5,"cache_write":5}}}},"iflowcn":{"id":"iflowcn","env":["IFLOW_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://apis.iflow.cn/v1","name":"iFlow","doc":"https://platform.iflow.cn/en/docs","models":{"qwen3-coder-plus":{"id":"qwen3-coder-plus","name":"Qwen3-Coder-Plus","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-01","last_updated":"2025-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0}},"deepseek-v3":{"id":"deepseek-v3","name":"DeepSeek-V3","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-10","release_date":"2024-12-26","last_updated":"2024-12-26","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32000},"cost":{"input":0,"output":0}},"qwen3-235b-a22b-thinking-2507":{"id":"qwen3-235b-a22b-thinking-2507","name":"Qwen3-235B-A22B-Thinking","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-01","last_updated":"2025-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0}},"glm-4.6":{"id":"glm-4.6","name":"GLM-4.6","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-10","release_date":"2024-12-01","last_updated":"2025-11-13","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":128000},"cost":{"input":0,"output":0}},"qwen3-235b-a22b-instruct":{"id":"qwen3-235b-a22b-instruct","name":"Qwen3-235B-A22B-Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-01","last_updated":"2025-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0}},"qwen3-235b":{"id":"qwen3-235b","name":"Qwen3-235B-A22B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-10","release_date":"2024-12-01","last_updated":"2024-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32000},"cost":{"input":0,"output":0}},"kimi-k2-0905":{"id":"kimi-k2-0905","name":"Kimi-K2-0905","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2025-09-05","last_updated":"2025-09-05","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0}},"qwen3-32b":{"id":"qwen3-32b","name":"Qwen3-32B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-10","release_date":"2024-12-01","last_updated":"2024-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32000},"cost":{"input":0,"output":0}},"qwen3-vl-plus":{"id":"qwen3-vl-plus","name":"Qwen3-VL-Plus","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2025-01-01","last_updated":"2025-01-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000},"cost":{"input":0,"output":0}},"qwen3-max-preview":{"id":"qwen3-max-preview","name":"Qwen3-Max-Preview","description":"Flagship Qwen model for complex reasoning, coding, and agentic workflows","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2025-01-01","last_updated":"2025-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000},"cost":{"input":0,"output":0}},"deepseek-r1":{"id":"deepseek-r1","name":"DeepSeek-R1","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2025-01-20","last_updated":"2025-01-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32000},"cost":{"input":0,"output":0}},"deepseek-v3.2":{"id":"deepseek-v3.2","name":"DeepSeek-V3.2-Exp","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2025-01-01","last_updated":"2025-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":64000},"cost":{"input":0,"output":0}},"qwen3-max":{"id":"qwen3-max","name":"Qwen3-Max","description":"Flagship Qwen model for complex reasoning, coding, and agentic workflows","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2025-01-01","last_updated":"2025-01-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000},"cost":{"input":0,"output":0}},"kimi-k2":{"id":"kimi-k2","name":"Kimi-K2","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-10","release_date":"2024-12-01","last_updated":"2024-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":64000},"cost":{"input":0,"output":0}}}},"scx-ai":{"id":"scx-ai","env":["SCX_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.scx.ai/v1","name":"SCX.ai","doc":"https://platform.scx.ai/docs","models":{"Qwen3.8-Max":{"id":"Qwen3.8-Max","name":"Qwen3.8 Max","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":983616,"output":131072},"cost":{"input":1.815,"output":5.4461,"cache_read":0.17,"cache_write":2.5}},"GLM-5.2":{"id":"GLM-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0.55,"output":1.784,"cache_read":0.111}},"gpt-oss-120b":{"id":"gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.17,"output":0.55}},"MiniMax-M2.7":{"id":"MiniMax-M2.7","name":"MiniMax-M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":196608,"output":196608},"cost":{"input":0.48,"output":1.79,"cache_read":0.05}}}},"evroc":{"id":"evroc","env":["EVROC_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://models.think.evroc.com/v1","name":"evroc","doc":"https://docs.evroc.com/products/think/overview.html","models":{"evroc/roc":{"id":"evroc/roc","name":"roc","description":"Multimodal reasoning model for visual analysis, planning, and tool use","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2026-01","release_date":"2026-06-06","last_updated":"2026-06-06","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":262144},"cost":{"input":2.875,"output":11.516}},"mistralai/Voxtral-Small-24B-2507":{"id":"mistralai/Voxtral-Small-24B-2507","name":"Voxtral Small 24B","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"voxtral","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2025-03-01","last_updated":"2025-03-01","modalities":{"input":["audio","text"],"output":["text"]},"open_weights":true,"limit":{"context":32000,"output":32000},"cost":{"input":0.0023,"output":0.0023,"output_audio":2.3}},"mistralai/Mistral-Medium-3.5-128B":{"id":"mistralai/Mistral-Medium-3.5-128B","name":"Mistral Medium 3.5","description":"Balanced Mistral model for enterprise assistants, multilingual work, and tools","family":"mistral-medium","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-29","last_updated":"2026-04-29","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":1.725,"output":6.9}},"nvidia/Llama-3.3-70B-Instruct-FP8":{"id":"nvidia/Llama-3.3-70B-Instruct-FP8","name":"Llama-3.3-70B-Instruct","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":1.15,"output":1.15}},"google/gemma-4-26B-A4B-it":{"id":"google/gemma-4-26B-A4B-it","name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.144,"output":0.575}},"zai-org/GLM-5.2":{"id":"zai-org/GLM-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":524288,"output":131072},"cost":{"input":1.4375,"output":5.75}},"Qwen/Qwen3.8-27B":{"id":"Qwen/Qwen3.8-27B","name":"Qwen3.8-27B","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.87,"output":3.5}},"Qwen/Qwen3-Reranker-4B":{"id":"Qwen/Qwen3-Reranker-4B","name":"Qwen3 Reranker 4B","description":"Reranking model for improving retrieval quality in search and recommendation systems","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2025-07-30","last_updated":"2025-07-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32000,"output":4096},"cost":{"input":0.0575,"output":0}},"Qwen/Qwen3.6-35B-A3B":{"id":"Qwen/Qwen3.6-35B-A3B","name":"Qwen3.6 35B-A3B","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.345,"output":1.38}},"Qwen/Qwen3-Embedding-8B":{"id":"Qwen/Qwen3-Embedding-8B","name":"Qwen3 Embedding 8B","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2025-07-30","last_updated":"2025-07-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":40960,"output":4096},"cost":{"input":0.115,"output":0.115}},"intfloat/multilingual-e5-large-instruct":{"id":"intfloat/multilingual-e5-large-instruct","name":"E5 Multi-Lingual Large Embeddings 0.6B","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2024-06-01","last_updated":"2024-06-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":512,"output":512},"cost":{"input":0.114,"output":0.114}},"KBLab/kb-whisper-large":{"id":"KBLab/kb-whisper-large","name":"KB Whisper","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"whisper","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2024-10-01","last_updated":"2024-10-01","modalities":{"input":["audio"],"output":["text"]},"open_weights":true,"limit":{"context":448,"output":448},"cost":{"input":0.0023,"output":0.0023,"output_audio":2.3}},"openai/whisper-large-v3":{"id":"openai/whisper-large-v3","name":"Whisper 3 Large","description":"Open Whisper checkpoint for robust multilingual transcription and captioning","family":"whisper","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2024-10-01","last_updated":"2024-10-01","modalities":{"input":["audio"],"output":["text"]},"open_weights":true,"limit":{"context":448,"output":4096},"cost":{"input":0.0023,"output":0.0023,"output_audio":2.3}},"openai/whisper-large-v3-turbo":{"id":"openai/whisper-large-v3-turbo","name":"Whisper Large v3 Turbo","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"whisper","attachment":false,"reasoning":false,"tool_call":false,"release_date":"2024-10-01","last_updated":"2024-10-01","modalities":{"input":["audio"],"output":["text"]},"open_weights":true,"limit":{"context":448,"output":448},"cost":{"input":0.0023,"output":0.0023,"output_audio":2.3}},"openai/gpt-oss-120b":{"id":"openai/gpt-oss-120b","name":"GPT OSS 120B","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"output":65536},"cost":{"input":0.23,"output":0.92}},"moonshotai/Kimi-K2.6":{"id":"moonshotai/Kimi-K2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":1.4375,"output":5.75}}}},"echo":{"id":"echo","env":["ECHO_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://echo.tracerml.ai/v1","name":"Echo","doc":"https://echo.tracerml.ai/docs/api","models":{"echo":{"id":"echo","name":"Echo","description":"Adaptive model for coding, reasoning, and tool-driven agent workflows through one OpenAI-compatible endpoint","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-07-19","last_updated":"2026-08-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":65536},"status":"beta","cost":{"input":10,"output":50}}}},"aixy":{"id":"aixy","env":["AIXY_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.aixy-gateway.com/v1","name":"Aixy","doc":"https://docs.aixy-gateway.com/integrations/overview","models":{"openai/gpt-4.1-mini":{"id":"openai/gpt-4.1-mini","name":"GPT-4.1 mini","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1}}}},"impossibl":{"id":"impossibl","env":["IMPOSSIBL_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.impossibl.com/v1","name":"Impossibl","doc":"https://impossibl.com/docs/models","models":{"qwen/qwen3.7-max":{"id":"qwen/qwen3.7-max","name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":2.5,"output":7.5,"cache_read":0.5}},"qwen/qwen3.8-max-preview":{"id":"qwen/qwen3.8-max-preview","name":"Qwen3.8 Max Preview","description":"Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]}],"tool_call":true,"temperature":true,"release_date":"2026-07-19","last_updated":"2026-07-19","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":2.5,"output":7.5}},"qwen/qwen3.6-flash":{"id":"qwen/qwen3.6-flash","name":"Qwen3.6 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-27","last_updated":"2026-04-27","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.05,"tiers":[{"input":1,"output":4,"cache_read":0.2,"tier":{"type":"context","size":262144}}],"context_over_200k":{"input":1,"output":4,"cache_read":0.2}}},"qwen/qwen3.7-plus":{"id":"qwen/qwen3.7-plus","name":"Qwen3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":0.4,"output":1.6,"cache_read":0.08,"tiers":[{"input":1.2,"output":4.8,"cache_read":0.24,"tier":{"type":"context","size":262144}}],"context_over_200k":{"input":1.2,"output":4.8,"cache_read":0.24}}},"groq/gpt-oss-20b":{"id":"groq/gpt-oss-20b","name":"GPT OSS 20B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.075,"output":0.3,"cache_read":0.0375}},"groq/gpt-oss-120b":{"id":"groq/gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"xiaomi/mimo-v2.5":{"id":"xiaomi/mimo-v2.5","name":"MiMo-V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.003}},"anthropic/claude-sonnet-4-6":{"id":"anthropic/claude-sonnet-4-6","name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024},{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"anthropic/claude-opus-4-5":{"id":"anthropic/claude-opus-4-5","name":"Claude Opus 4.5 (latest)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-opus-4-6":{"id":"anthropic/claude-opus-4-6","name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024},{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-opus-4-7":{"id":"anthropic/claude-opus-4-7","name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-fable-5":{"id":"anthropic/claude-fable-5","name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"anthropic/claude-haiku-4-5":{"id":"anthropic/claude-haiku-4-5","name":"Claude Haiku 4.5 (latest)","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"anthropic/claude-sonnet-4-5":{"id":"anthropic/claude-sonnet-4-5","name":"Claude Sonnet 4.5 (latest)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"anthropic/claude-opus-4-8":{"id":"anthropic/claude-opus-4-8","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-sonnet-5":{"id":"anthropic/claude-sonnet-5","name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"google/gemini-3.1-pro-preview":{"id":"google/gemini-3.1-pro-preview","name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"google/gemini-2.5-flash-lite":{"id":"google/gemini-2.5-flash-lite","name":"Gemini 2.5 Flash-Lite","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":512,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01}},"google/gemini-3.6-flash":{"id":"google/gemini-3.6-flash","name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":7.5,"cache_read":0.15}},"google/gemini-3.1-flash-lite":{"id":"google/gemini-3.1-flash-lite","name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025}},"google/gemini-3.5-flash":{"id":"google/gemini-3.5-flash","name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15}},"google/gemini-3.5-flash-lite":{"id":"google/gemini-3.5-flash-lite","name":"Gemini 3.5 Flash Lite","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"google/gemini-2.5-pro":{"id":"google/gemini-2.5-pro","name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}}},"google/gemini-2.5-flash":{"id":"google/gemini-2.5-flash","name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"thinkingmachines/inkling":{"id":"thinkingmachines/inkling","name":"Inkling","description":"Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio","family":"ling","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"output":65536},"cost":{"input":1.87,"output":4.68,"cache_read":0.374}},"deepseek/deepseek-v4-flash":{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.19,"output":0.51,"cache_read":0.028}},"deepseek/deepseek-v4-pro":{"id":"deepseek/deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":1.74,"output":3.48,"cache_read":0.145}},"openai/gpt-5-nano":{"id":"openai/gpt-5-nano","name":"GPT-5 Nano","description":"Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005}},"openai/gpt-4.1-nano":{"id":"openai/gpt-4.1-nano","name":"GPT-4.1 nano","description":"Tiny GPT-4.1 option for classification, routing, and very high-volume tasks","family":"gpt-nano","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.025}},"openai/gpt-5-codex":{"id":"openai/gpt-5-codex","name":"GPT-5-Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-09-15","last_updated":"2025-09-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/gpt-5-pro":{"id":"openai/gpt-5-pro","name":"GPT-5 Pro","description":"Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-10-06","last_updated":"2025-10-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":272000},"cost":{"input":15,"output":120}},"openai/gpt-5.1-codex-mini":{"id":"openai/gpt-5.1-codex-mini","name":"GPT-5.1 Codex mini","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"openai/gpt-5.1-codex":{"id":"openai/gpt-5.1-codex","name":"GPT-5.1 Codex","description":"Codex GPT for repository edits, code review, and practical software agents","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/gpt-5.6-sol":{"id":"openai/gpt-5.6-sol","name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":45,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1,"cache_write":12.5}}},"openai/gpt-5.2-codex":{"id":"openai/gpt-5.2-codex","name":"GPT-5.2 Codex","description":"Code-specialist GPT for repository edits, reviews, and long-running software agents","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-4.1-mini":{"id":"openai/gpt-4.1-mini","name":"GPT-4.1 mini","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1}},"openai/gpt-5.4":{"id":"openai/gpt-5.4","name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}}},"openai/gpt-4-turbo":{"id":"openai/gpt-4-turbo","name":"GPT-4 Turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-12","release_date":"2023-11-06","last_updated":"2024-04-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096},"cost":{"input":10,"output":30}},"openai/gpt-5.1":{"id":"openai/gpt-5.1","name":"GPT-5.1","description":"Sharper GPT-5 generation for coding, product work, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/o1":{"id":"openai/o1","name":"o1","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5}},"openai/gpt-4o":{"id":"openai/gpt-4o","name":"GPT-4o","description":"Omni-era GPT for multimodal chat, practical coding, and general assistants","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"openai/gpt-5.6-luna":{"id":"openai/gpt-5.6-luna","name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}}},"openai/gpt-5.3-codex":{"id":"openai/gpt-5.3-codex","name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-4o-mini":{"id":"openai/gpt-4o-mini","name":"GPT-4o mini","description":"Small omni GPT for cheap multimodal assistance and production-scale traffic","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"openai/gpt-4.1":{"id":"openai/gpt-4.1","name":"GPT-4.1","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1047576,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5}},"openai/gpt-5.4-nano":{"id":"openai/gpt-5.4-nano","name":"GPT-5.4 nano","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02}},"openai/gpt-5.5-pro":{"id":"openai/gpt-5.5-pro","name":"GPT-5.5 Pro","description":"Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"cache_read":3,"tiers":[{"input":60,"output":270,"cache_read":3,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270,"cache_read":3}}},"openai/gpt-5.4-mini":{"id":"openai/gpt-5.4-mini","name":"GPT-5.4 mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075}},"openai/gpt-3.5-turbo":{"id":"openai/gpt-3.5-turbo","name":"GPT-3.5-turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2021-09-01","release_date":"2023-03-01","last_updated":"2023-11-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":16385,"output":4096},"cost":{"input":0.5,"output":1.5}},"openai/gpt-5-mini":{"id":"openai/gpt-5-mini","name":"GPT-5 Mini","description":"Small GPT-5 for responsive agents, coding help, and everyday automation","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"openai/gpt-5.4-pro":{"id":"openai/gpt-5.4-pro","name":"GPT-5.4 Pro","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"cache_read":3,"tiers":[{"input":60,"output":270,"cache_read":3,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270,"cache_read":3}}},"openai/gpt-5.6-terra":{"id":"openai/gpt-5.6-terra","name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}}},"openai/gpt-5.2":{"id":"openai/gpt-5.2","name":"GPT-5.2","description":"Reliable GPT generation for broad coding, writing, and tool-assisted product work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-5":{"id":"openai/gpt-5","name":"GPT-5","description":"Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/o4-mini":{"id":"openai/o4-mini","name":"o4-mini","description":"Fast o-series model for compact reasoning, coding, and tool use","family":"o-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.275}},"openai/o3-mini":{"id":"openai/o3-mini","name":"o3-mini","description":"Smaller o-series reasoner for economical coding, math, and planning tasks","family":"o-mini","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai/o3":{"id":"openai/o3","name":"o3","description":"Deliberate o-series reasoner for hard math, coding, and multi-step analysis","family":"o","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":0.5}},"openai/gpt-5.5":{"id":"openai/gpt-5.5","name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}}},"moonshotai/kimi-k3":{"id":"moonshotai/kimi-k3","name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3}},"xai/grok-4.3":{"id":"xai/grok-4.3","name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2}},"xai/grok-4.20-0309-reasoning":{"id":"xai/grok-4.20-0309-reasoning","name":"Grok 4.20 (Reasoning)","description":"Reasoning Grok for document-heavy analysis and long-horizon tool use","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-03-09","last_updated":"2026-03-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2}},"xai/grok-4.5":{"id":"xai/grok-4.5","name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-08","last_updated":"2026-07-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.3}},"xai/grok-build-0.1":{"id":"xai/grok-build-0.1","name":"Grok Build 0.1","description":"Fast Grok coding model tuned for agentic engineering and iterative edits","family":"grok-build","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":1,"output":2,"cache_read":0.2}},"xai/grok-4.20-0309-non-reasoning":{"id":"xai/grok-4.20-0309-non-reasoning","name":"Grok 4.20 (Non-Reasoning)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-03-09","last_updated":"2026-03-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2}},"zai/glm-4.7":{"id":"zai/glm-4.7","name":"GLM-4.7","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"zai/glm-4.5-air":{"id":"zai/glm-4.5-air","name":"GLM-4.5-Air","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":98304},"cost":{"input":0.2,"output":1.1,"cache_read":0.03}},"zai/glm-4.6":{"id":"zai/glm-4.6","name":"GLM-4.6","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"zai/glm-5.2":{"id":"zai/glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"zai/glm-4.5":{"id":"zai/glm-4.5","name":"GLM-4.5","description":"Hybrid-reasoning GLM release that made the 4.5 line broadly useful","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":98304},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"zai/glm-5":{"id":"zai/glm-5","name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"zai/glm-5.1":{"id":"zai/glm-5.1","name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"zai/glm-5-turbo":{"id":"zai/glm-5-turbo","name":"GLM-5-Turbo","description":"Faster GLM-5 lane for coding agents that need lower latency","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-03-16","last_updated":"2026-03-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24}},"fireworks/glm-5.2":{"id":"fireworks/glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.14}},"fireworks/gpt-oss-20b":{"id":"fireworks/gpt-oss-20b","name":"GPT OSS 20B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.07,"output":0.3,"cache_read":0.035}},"fireworks/gpt-oss-120b":{"id":"fireworks/gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":0.6,"cache_read":0.015}},"cerebras/gpt-oss-120b":{"id":"cerebras/gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.35,"output":0.75}}}},"llmgateway-providers":{"id":"llmgateway-providers","env":["LLMGATEWAY_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.llmgateway.io/v1","name":"LLM Gateway","doc":"https://llmgateway.io/docs","models":{"vertex-openai/glm-4.7":{"id":"vertex-openai/glm-4.7","name":"GLM-4.7 (Vertex AI (OpenAI-compatible))","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":128000},"cost":{"input":0.6,"output":2.2}},"vertex-openai/qwen3-next-80b-a3b-thinking":{"id":"vertex-openai/qwen3-next-80b-a3b-thinking","name":"Qwen3 Next 80B A3B Thinking (Vertex AI (OpenAI-compatible))","description":"Efficient Qwen thinking model for local reasoning, math, and coding agents","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":1.2}},"vertex-openai/qwen3-next-80b-a3b-instruct":{"id":"vertex-openai/qwen3-next-80b-a3b-instruct","name":"Qwen3 Next 80B A3B Instruct (Vertex AI (OpenAI-compatible))","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":1.2}},"vertex-openai/kimi-k2-thinking":{"id":"vertex-openai/kimi-k2-thinking","name":"Kimi K2 Thinking (Vertex AI (OpenAI-compatible))","description":"Thinking Kimi model for slower research passes, planning, and hard technical questions","family":"kimi-thinking","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2024-08","release_date":"2025-11-06","last_updated":"2025-11-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.6,"output":2.5,"cache_read":0.06}},"vertex-openai/deepseek-v3.2":{"id":"vertex-openai/deepseek-v3.2","name":"DeepSeek V3.2 (Vertex AI (OpenAI-compatible))","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"output":65536},"cost":{"input":0.56,"output":1.68,"cache_read":0.056}},"vertex-openai/glm-5":{"id":"vertex-openai/glm-5","name":"GLM-5 (Vertex AI (OpenAI-compatible))","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":32768},"cost":{"input":1,"output":3.2,"cache_read":0.1}},"vertex-openai/qwen3-235b-a22b-instruct-2507":{"id":"vertex-openai/qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507 (Vertex AI (OpenAI-compatible))","description":"Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-21","last_updated":"2025-07-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.22,"output":0.88}},"vertex-openai/grok-4-6":{"id":"vertex-openai/grok-4-6","name":"Grok 4.6 (Vertex AI (OpenAI-compatible))","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5}},"vertex-openai/qwen3-coder-480b-a35b-instruct":{"id":"vertex-openai/qwen3-coder-480b-a35b-instruct","name":"Qwen3 Coder 480B A35B Instruct (Vertex AI (OpenAI-compatible))","description":"Open Qwen coding heavyweight for repository reasoning and agentic engineering","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.22,"output":1.8,"cache_read":0.022}},"vertex-openai/grok-4-20-non-reasoning":{"id":"vertex-openai/grok-4-20-non-reasoning","name":"Grok 4.20 Non-Reasoning (Vertex AI (OpenAI-compatible))","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"grok","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-09","last_updated":"2026-03-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"vertex-openai/grok-4-20-reasoning":{"id":"vertex-openai/grok-4-20-reasoning","name":"Grok 4.20 Reasoning (Vertex AI (OpenAI-compatible))","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-09","last_updated":"2026-03-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"baidu/kimi-k2.6":{"id":"baidu/kimi-k2.6","name":"Kimi K2.6 (Baidu)","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.16}},"baidu/glm-5.2":{"id":"baidu/glm-5.2","name":"GLM-5.2 (Baidu)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"baidu/deepseek-v4-flash":{"id":"baidu/deepseek-v4-flash","name":"DeepSeek V4 Flash (Baidu)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.44,"output":1.32,"cache_read":0.044}},"baidu/glm-5":{"id":"baidu/glm-5","name":"GLM-5 (Baidu)","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"baidu/glm-5.1":{"id":"baidu/glm-5.1","name":"GLM-5.1 (Baidu)","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"baidu/deepseek-v4-pro":{"id":"baidu/deepseek-v4-pro","name":"DeepSeek V4 Pro (Baidu)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":1.32,"output":3.96,"cache_read":0.132}},"baidu/glm-5.3":{"id":"baidu/glm-5.3","name":"GLM-5.3 (Baidu)","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"aws-mantle/gpt-5.6-sol":{"id":"aws-mantle/gpt-5.6-sol","name":"GPT-5.6 Sol (AWS Mantle)","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":278528,"output":128000},"cost":{"input":5.5,"output":33,"cache_read":0.55,"cache_write":6.875}},"aws-mantle/gpt-5.6-luna":{"id":"aws-mantle/gpt-5.6-luna","name":"GPT-5.6 Luna (AWS Mantle)","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":278528,"output":128000},"cost":{"input":0.22,"output":1.32,"cache_read":0.022,"cache_write":0.275}},"aws-mantle/gpt-5.6-terra":{"id":"aws-mantle/gpt-5.6-terra","name":"GPT-5.6 Terra (AWS Mantle)","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":278528,"output":128000},"cost":{"input":2.2,"output":13.2,"cache_read":0.22,"cache_write":2.75}},"gonka24/minimax-m2.7":{"id":"gonka24/minimax-m2.7","name":"MiniMax M2.7 (Gonka24)","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131100},"cost":{"input":0.08,"output":0.32,"cache_read":0.017}},"gonka24/kimi-k2.6":{"id":"gonka24/kimi-k2.6","name":"Kimi K2.6 (Gonka24)","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":98304},"cost":{"input":0.22,"output":1.137,"cache_read":0.048}},"gonka24/deepseek-v4-flash":{"id":"gonka24/deepseek-v4-flash","name":"DeepSeek V4 Flash (Gonka24)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":390000,"output":16384},"cost":{"input":0.051,"output":0.104,"cache_read":0.0097}},"embercloud/glm-4.7":{"id":"embercloud/glm-4.7","name":"GLM-4.7 (EmberCloud)","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":131000},"cost":{"input":0.38,"output":1.98,"cache_read":0.19}},"embercloud/glm-4.5-air":{"id":"embercloud/glm-4.5-air","name":"GLM-4.5 Air (EmberCloud)","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131000,"output":96000},"cost":{"input":0.13,"output":0.85,"cache_read":0.025}},"embercloud/glm-5.2":{"id":"embercloud/glm-5.2","name":"GLM-5.2 (EmberCloud)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":203000,"output":131000},"cost":{"input":1.26,"output":3.96,"cache_read":0.234}},"embercloud/qwen3-coder-next":{"id":"embercloud/qwen3-coder-next","name":"Qwen3 Coder Next (EmberCloud)","description":"Open-weight Qwen coding model for agents, repository edits, and multi-turn tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-09","release_date":"2026-02-03","last_updated":"2026-02-03","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.108,"output":0.675,"cache_read":0.06}},"embercloud/glm-4.5":{"id":"embercloud/glm-4.5","name":"GLM-4.5 (EmberCloud)","description":"Hybrid-reasoning GLM release that made the 4.5 line broadly useful","family":"glm","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131000,"output":96000},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"embercloud/glm-5":{"id":"embercloud/glm-5","name":"GLM-5 (EmberCloud)","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":203000,"output":131000},"cost":{"input":0.72,"output":2.3,"cache_read":0.144}},"embercloud/kimi-k2.5":{"id":"embercloud/kimi-k2.5","name":"Kimi K2.5 (EmberCloud)","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.405,"output":1.98,"cache_read":0.225}},"embercloud/glm-5.1":{"id":"embercloud/glm-5.1","name":"GLM-5.1 (EmberCloud)","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":203000,"output":131000},"cost":{"input":0.931,"output":2.93,"cache_read":0.173}},"embercloud/glm-4.7-flash":{"id":"embercloud/glm-4.7-flash","name":"GLM-4.7 Flash (EmberCloud)","description":"Budget GLM lane for fast coding help, routing, and everyday automation","family":"glm-flash","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":131000},"cost":{"input":0.06,"output":0.4,"cache_read":0.01}},"scx-ai/minimax-m2.7":{"id":"scx-ai/minimax-m2.7","name":"MiniMax M2.7 (SCX.ai (Turbo))","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":196608,"output":196608},"cost":{"input":0.48,"output":1.79,"cache_read":0.05}},"scx-ai/qwen3-32b":{"id":"scx-ai/qwen3-32b","name":"Qwen3 32B (SCX.ai (Turbo))","description":"Dense open Qwen model for self-hosted chat, reasoning, and coding","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":32768,"output":8192},"cost":{"input":0.36,"output":0.87}},"scx-ai/llama-4-maverick-17b-instruct":{"id":"scx-ai/llama-4-maverick-17b-instruct","name":"Llama 4 Maverick 17B Instruct (SCX.ai (Turbo))","description":"Open multimodal Llama for strong reasoning with efficient everyday serving","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":8192},"cost":{"input":0.53,"output":1.62}},"scx-ai/gemma-4-31b-it":{"id":"scx-ai/gemma-4-31b-it","name":"Gemma 4 31B IT (SCX.ai (Turbo))","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":8192},"cost":{"input":0.3,"output":0.91}},"scx-ai/gpt-oss-120b":{"id":"scx-ai/gpt-oss-120b","name":"GPT OSS 120B (SCX.ai (Turbo))","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.17,"output":0.55}},"groq/gpt-oss-20b":{"id":"groq/gpt-oss-20b","name":"GPT OSS 20B (Groq)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32766},"cost":{"input":0.1,"output":0.5}},"groq/gpt-oss-120b":{"id":"groq/gpt-oss-120b","name":"GPT OSS 120B (Groq)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32766},"cost":{"input":0.15,"output":0.75}},"google-vertex/gemini-3.1-pro-preview":{"id":"google-vertex/gemini-3.1-pro-preview","name":"Gemini 3.1 Pro (Preview) (Google Vertex AI)","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"google-vertex/gemini-2.5-flash-lite":{"id":"google-vertex/gemini-2.5-flash-lite","name":"Gemini 2.5 Flash Lite (Google Vertex AI)","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65535},"cost":{"input":0.1,"output":0.4,"cache_read":0.01}},"google-vertex/gemini-3.6-flash":{"id":"google-vertex/gemini-3.6-flash","name":"Gemini 3.6 Flash (Google Vertex AI)","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.08333}},"google-vertex/gemini-3.1-flash-lite":{"id":"google-vertex/gemini-3.1-flash-lite","name":"Gemini 3.1 Flash Lite (Google Vertex AI)","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"cache_write":0.08333}},"google-vertex/gemini-3.5-flash":{"id":"google-vertex/gemini-3.5-flash","name":"Gemini 3.5 Flash (Google Vertex AI)","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"cache_write":0.08333}},"google-vertex/gemini-3.5-flash-lite":{"id":"google-vertex/gemini-3.5-flash-lite","name":"Gemini 3.5 Flash Lite (Google Vertex AI)","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"cache_write":0.08333}},"google-vertex/gemini-3-flash-preview":{"id":"google-vertex/gemini-3-flash-preview","name":"Gemini 3 Flash (Preview) (Google Vertex AI)","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65535},"cost":{"input":0.5,"output":3,"cache_read":0.05}},"google-vertex/gemini-3.8-flash":{"id":"google-vertex/gemini-3.8-flash","name":"Gemini 3.8 Flash (Google Vertex AI)","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.08333}},"google-vertex/gemini-3.7-flash":{"id":"google-vertex/gemini-3.7-flash","name":"Gemini 3.7 Flash (Google Vertex AI)","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.08333}},"google-vertex/gemini-2.5-pro":{"id":"google-vertex/gemini-2.5-pro","name":"Gemini 2.5 Pro (Google Vertex AI)","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]},{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}}},"google-vertex/gemini-2.5-flash":{"id":"google-vertex/gemini-2.5-flash","name":"Gemini 2.5 Flash (Google Vertex AI)","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1,"max":24576},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65535},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"quartz/gemini-3.1-pro-preview":{"id":"quartz/gemini-3.1-pro-preview","name":"Gemini 3.1 Pro (Preview) (Quartz)","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"vertex-anthropic/claude-sonnet-4-6":{"id":"vertex-anthropic/claude-sonnet-4-6","name":"Claude Sonnet 4.6 (Vertex AI (Anthropic))","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"vertex-anthropic/claude-opus-4-6":{"id":"vertex-anthropic/claude-opus-4-6","name":"Claude Opus 4.6 (Vertex AI (Anthropic))","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"vertex-anthropic/claude-opus-4-7":{"id":"vertex-anthropic/claude-opus-4-7","name":"Claude Opus 4.7 (Vertex AI (Anthropic))","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5}},"vertex-anthropic/claude-haiku-4-5":{"id":"vertex-anthropic/claude-haiku-4-5","name":"Claude Haiku 4.5 (Vertex AI (Anthropic))","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1}},"vertex-anthropic/claude-sonnet-4-5":{"id":"vertex-anthropic/claude-sonnet-4-5","name":"Claude Sonnet 4.5 (Vertex AI (Anthropic))","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3}},"vertex-anthropic/claude-sonnet-5":{"id":"vertex-anthropic/claude-sonnet-5","name":"Claude Sonnet 5 (Vertex AI (Anthropic))","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"vertex-anthropic/claude-opus-4-5-20251101":{"id":"vertex-anthropic/claude-opus-4-5-20251101","name":"Claude Opus 4.5 (Vertex AI (Anthropic))","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":31999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2025-11-01","last_updated":"2025-11-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5}},"xiaomi/mimo-v2.5":{"id":"xiaomi/mimo-v2.5","name":"MiMo V2.5 (Xiaomi)","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028,"tiers":[{"input":0.8,"output":4,"cache_read":0.16,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.8,"output":4,"cache_read":0.16}}},"xiaomi/mimo-v2.5-pro":{"id":"xiaomi/mimo-v2.5-pro","name":"MiMo V2.5 Pro (Xiaomi)","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0.435,"output":0.87,"cache_read":0.0036,"tiers":[{"input":2,"output":6,"cache_read":0.4,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2,"output":6,"cache_read":0.4}}},"minimax/minimax-m2.1-lightning":{"id":"minimax/minimax-m2.1-lightning","name":"MiniMax M2.1 Lightning (MiniMax)","description":"High-speed MiniMax model for low-latency coding and agent workflows","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":196608,"output":131072},"cost":{"input":0.12,"output":0.48}},"minimax/minimax-m2.1":{"id":"minimax/minimax-m2.1","name":"MiniMax M2.1 (MiniMax)","description":"Earlier MiniMax agent model for practical coding and productivity tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":196608,"output":131072},"cost":{"input":0.27,"output":1.1}},"minimax/minimax-m2":{"id":"minimax/minimax-m2","name":"MiniMax M2 (MiniMax)","description":"Efficient open MiniMax model built for coding agents and tool-heavy workflows","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-10-27","last_updated":"2025-10-27","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":196608,"output":131072},"cost":{"input":0.2,"output":1,"cache_read":0.03}},"minimax/minimax-m2.7-highspeed":{"id":"minimax/minimax-m2.7-highspeed","name":"MiniMax M2.7 Highspeed (MiniMax)","description":"Low-latency M2.7 variant for interactive coding plans and agent loops","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131100},"cost":{"input":0.6,"output":2.4,"cache_read":0.06}},"minimax/minimax-m2.7":{"id":"minimax/minimax-m2.7","name":"MiniMax M2.7 (MiniMax)","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131100},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"minimax/minimax-m2.5":{"id":"minimax/minimax-m2.5","name":"MiniMax M2.5 (MiniMax)","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131100},"cost":{"input":0.3,"output":1.2,"cache_read":0.03}},"minimax/minimax-m3":{"id":"minimax/minimax-m3","name":"MiniMax M3 (MiniMax)","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-06-01","last_updated":"2026-06-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":512000,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_read":0.12}},"minimax/minimax-m2.5-highspeed":{"id":"minimax/minimax-m2.5-highspeed","name":"MiniMax M2.5 Highspeed (MiniMax)","description":"High-speed MiniMax model for low-latency coding and agent workflows","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-13","last_updated":"2026-02-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131100},"cost":{"input":0.6,"output":2.4,"cache_read":0.03}},"minimax/minimax-text-01":{"id":"minimax/minimax-text-01","name":"MiniMax Text 01 (MiniMax)","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-01-15","last_updated":"2025-01-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0.2,"output":1.1}},"alibaba/qwen3.7-max":{"id":"alibaba/qwen3.7-max","name":"Qwen3.7 Max (Alibaba Cloud)","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":2.5,"output":7.5,"cache_read":0.5,"cache_write":3.125}},"alibaba/qwen3-coder-plus":{"id":"alibaba/qwen3-coder-plus","name":"Qwen3 Coder Plus (Alibaba Cloud)","description":"Hosted Qwen coder for software agents, repo edits, and long-context code","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-23","last_updated":"2025-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":66000},"cost":{"input":1,"output":5,"cache_read":0.2,"cache_write":1.25}},"alibaba/qwen35-397b-a17b":{"id":"alibaba/qwen35-397b-a17b","name":"Qwen3.5 397B A17B (Alibaba Cloud)","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-16","last_updated":"2026-02-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":65536},"cost":{"input":0.6,"output":3.6}},"alibaba/qwen3-coder-flash":{"id":"alibaba/qwen3-coder-flash","name":"Qwen3 Coder Flash (Alibaba Cloud)","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.3,"output":1.5,"cache_read":0.06,"cache_write":0.375}},"alibaba/qwen-max":{"id":"alibaba/qwen-max","name":"Qwen Max (Alibaba Cloud)","description":"Flagship Qwen model for complex reasoning, coding, and agentic workflows","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2024-04-03","last_updated":"2025-01-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"output":8192},"cost":{"input":1.6,"output":6.4}},"alibaba/qwen3.6-plus":{"id":"alibaba/qwen3.6-plus","name":"Qwen3.6 Plus (Alibaba Cloud)","description":"Earlier Qwen multimodal workhorse for million-token agent and document tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"cache_write":0.625,"tiers":[{"input":2,"output":6,"cache_read":0.2,"cache_write":2.5,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2,"output":6,"cache_read":0.2,"cache_write":2.5}}},"alibaba/qwen-flash":{"id":"alibaba/qwen-flash","name":"Qwen Flash (Alibaba Cloud)","description":"Efficient Qwen model for fast chat, extraction, and high-volume workloads","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32000},"cost":{"input":0.05,"output":0.4,"cache_read":0.01,"cache_write":0.0625}},"alibaba/glm-5.2":{"id":"alibaba/glm-5.2","name":"GLM-5.2 (Alibaba Cloud)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.28}},"alibaba/deepseek-v4-flash":{"id":"alibaba/deepseek-v4-flash","name":"DeepSeek V4 Flash (Alibaba Cloud)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":393216},"cost":{"input":0.2,"output":0.4,"cache_read":0.04}},"alibaba/qwen3-vl-plus":{"id":"alibaba/qwen3-vl-plus","name":"Qwen3 VL Plus (Alibaba Cloud)","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":32768},"cost":{"input":0.2,"output":1.6,"cache_read":0.04,"cache_write":0.25}},"alibaba/qwen-coder-plus":{"id":"alibaba/qwen-coder-plus","name":"Qwen Coder Plus (Alibaba Cloud)","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2024-09-18","last_updated":"2024-09-18","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":8192},"cost":{"input":0.502,"output":1.004}},"alibaba/qwen3.7-flash":{"id":"alibaba/qwen3.7-flash","name":"Qwen3.7 Flash (Alibaba Cloud)","description":"Lightweight multimodal Qwen model for high-throughput text, image, and video tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":991000,"output":65536},"cost":{"input":0.03,"output":0.13,"cache_read":0.006,"cache_write":0.0375}},"alibaba/qwen3.6-35b-a3b":{"id":"alibaba/qwen3.6-35b-a3b","name":"Qwen3.6 35B A3B (Alibaba Cloud)","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.375,"output":2.25}},"alibaba/qwen3-max":{"id":"alibaba/qwen3-max","name":"Qwen3 Max (Alibaba Cloud)","description":"Flagship Qwen3 model for coding agents, complex reasoning, and tool use","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32800},"cost":{"input":1.2,"output":6,"cache_read":0.24,"cache_write":1.5}},"alibaba/qwen3-vl-flash":{"id":"alibaba/qwen3-vl-flash","name":"Qwen3 VL Flash (Alibaba Cloud)","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":32768},"cost":{"input":0.05,"output":0.4,"cache_read":0.01}},"alibaba/qwen-plus":{"id":"alibaba/qwen-plus","name":"Qwen Plus (Alibaba Cloud)","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2024-01-25","last_updated":"2025-09-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":32000},"cost":{"input":0.4,"output":1.2,"cache_read":0.08,"cache_write":0.5}},"alibaba/qwen3.6-flash":{"id":"alibaba/qwen3.6-flash","name":"Qwen3.6 Flash (Alibaba Cloud)","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-27","last_updated":"2026-04-27","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.05,"cache_write":0.3125}},"alibaba/qwen3.8-flash":{"id":"alibaba/qwen3.8-flash","name":"Qwen3.8 Flash (Alibaba Cloud)","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.47,"cache_read":0.016,"cache_write":0.2}},"alibaba/qwen3.6-max-preview":{"id":"alibaba/qwen3.6-max-preview","name":"Qwen3.6 Max Preview (Alibaba Cloud)","description":"Flagship Qwen model for complex reasoning, coding, and agentic workflows","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-20","last_updated":"2026-04-20","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":65536},"cost":{"input":1.3,"output":7.8,"cache_read":0.13}},"alibaba/glm-5":{"id":"alibaba/glm-5","name":"GLM-5 (Alibaba Cloud)","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":16384},"cost":{"input":0.573,"output":2.58}},"alibaba/qwen3.8-max":{"id":"alibaba/qwen3.8-max","name":"Qwen3.8 Max (Alibaba Cloud)","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25,"cache_write":2.5}},"alibaba/kimi-k2.5":{"id":"alibaba/kimi-k2.5","name":"Kimi K2.5 (Alibaba Cloud)","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":98304},"cost":{"input":0.574,"output":3.011}},"alibaba/qwen3.7-plus":{"id":"alibaba/qwen3.7-plus","name":"Qwen3.7 Plus (Alibaba Cloud)","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.4,"output":1.6,"cache_read":0.08,"cache_write":0.5}},"alibaba/qwen-omni-turbo":{"id":"alibaba/qwen-omni-turbo","name":"Qwen Omni Turbo (Alibaba Cloud)","description":"Qwen omni model for text, vision, audio, and multimodal agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-04","release_date":"2025-01-19","last_updated":"2025-03-26","modalities":{"input":["text","image","audio","video"],"output":["text","audio"]},"open_weights":false,"limit":{"context":32768,"output":8192},"cost":{"input":0.2,"output":0.8}},"alibaba/deepseek-v4-pro":{"id":"alibaba/deepseek-v4-pro","name":"DeepSeek V4 Pro (Alibaba Cloud)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":393216},"cost":{"input":2.4,"output":4.8,"cache_read":0.2}},"alibaba/qwen-plus-latest":{"id":"alibaba/qwen-plus-latest","name":"Qwen Plus Latest (Alibaba Cloud)","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2024-09-09","last_updated":"2024-09-09","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32000},"cost":{"input":0.4,"output":1.2,"cache_read":0.08,"cache_write":0.5}},"scx-ai-gp/glm-5.2":{"id":"scx-ai-gp/glm-5.2","name":"GLM-5.2 (SCX.ai)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0.55,"output":1.784,"cache_read":0.111}},"scx-ai-gp/kimi-k2.7-code":{"id":"scx-ai-gp/kimi-k2.7-code","name":"Kimi K2.7 Code (SCX.ai)","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.89,"output":3.71,"cache_read":0.18}},"scx-ai-gp/kimi-k3":{"id":"scx-ai-gp/kimi-k3","name":"Kimi K3 (SCX.ai)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":2.83,"output":14.13,"cache_read":0.28}},"scx-ai-gp/glm-5.2-fast":{"id":"scx-ai-gp/glm-5.2-fast","name":"GLM-5.2 Turbo (SCX.ai)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.99,"output":6.16,"cache_read":0.4}},"scx-ai-gp/glm-5.3-flash":{"id":"scx-ai-gp/glm-5.3-flash","name":"GLM-5.3 Flash (SCX.ai)","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":0.13,"output":0.4,"cache_read":0.024}},"scx-ai-gp/qwen3.8-max":{"id":"scx-ai-gp/qwen3.8-max","name":"Qwen3.8 Max (SCX.ai)","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":1.815,"output":5.4461,"cache_read":0.21,"cache_write":2.5}},"scx-ai-gp/glm-5.3":{"id":"scx-ai-gp/glm-5.3","name":"GLM-5.3 (SCX.ai)","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":128000},"cost":{"input":1.3,"output":4,"cache_read":0.25}},"aws-bedrock/claude-sonnet-4-6":{"id":"aws-bedrock/claude-sonnet-4-6","name":"Claude Sonnet 4.6 (AWS Bedrock)","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"aws-bedrock/llama-4-scout-17b-instruct":{"id":"aws-bedrock/llama-4-scout-17b-instruct","name":"Llama 4 Scout 17B Instruct (AWS Bedrock)","description":"Open Llama with long-context vision for efficient multimodal agents","family":"llama","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"output":2048},"cost":{"input":0.17,"output":0.66}},"aws-bedrock/claude-opus-5":{"id":"aws-bedrock/claude-opus-5","name":"Claude Opus 5 (AWS Bedrock)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"aws-bedrock/claude-opus-4-1-20250805":{"id":"aws-bedrock/claude-opus-4-1-20250805","name":"Claude Opus 4.1 (AWS Bedrock)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":31999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75}},"aws-bedrock/claude-opus-4-6":{"id":"aws-bedrock/claude-opus-4-6","name":"Claude Opus 4.6 (AWS Bedrock)","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"aws-bedrock/claude-sonnet-4-5-20250929":{"id":"aws-bedrock/claude-sonnet-4-5-20250929","name":"Claude Sonnet 4.5 (2025-09-29) (AWS Bedrock)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"aws-bedrock/claude-opus-4-7":{"id":"aws-bedrock/claude-opus-4-7","name":"Claude Opus 4.7 (AWS Bedrock)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"aws-bedrock/claude-haiku-4-5-20251001":{"id":"aws-bedrock/claude-haiku-4-5-20251001","name":"Claude Haiku 4.5 (2025-10-01) (AWS Bedrock)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"aws-bedrock/claude-fable-5":{"id":"aws-bedrock/claude-fable-5","name":"Claude Fable 5 (AWS Bedrock)","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"aws-bedrock/llama-4-maverick-17b-instruct":{"id":"aws-bedrock/llama-4-maverick-17b-instruct","name":"Llama 4 Maverick 17B Instruct (AWS Bedrock)","description":"Open multimodal Llama for strong reasoning with efficient everyday serving","family":"llama","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"output":2048},"cost":{"input":0.24,"output":0.97}},"aws-bedrock/grok-4-3":{"id":"aws-bedrock/grok-4-3","name":"Grok 4.3 (AWS Bedrock)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-30","last_updated":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"aws-bedrock/claude-haiku-4-5":{"id":"aws-bedrock/claude-haiku-4-5","name":"Claude Haiku 4.5 (AWS Bedrock)","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"aws-bedrock/claude-sonnet-4-5":{"id":"aws-bedrock/claude-sonnet-4-5","name":"Claude Sonnet 4.5 (AWS Bedrock)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"aws-bedrock/llama-3.1-70b-instruct":{"id":"aws-bedrock/llama-3.1-70b-instruct","name":"Llama 3.1 70B Instruct (AWS Bedrock)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2024-07-23","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":2048},"cost":{"input":0.72,"output":0.72}},"aws-bedrock/grok-4-6":{"id":"aws-bedrock/grok-4-6","name":"Grok 4.6 (AWS Bedrock)","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5}},"aws-bedrock/claude-opus-4-8":{"id":"aws-bedrock/claude-opus-4-8","name":"Claude Opus 4.8 (AWS Bedrock)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"aws-bedrock/claude-sonnet-5":{"id":"aws-bedrock/claude-sonnet-5","name":"Claude Sonnet 5 (AWS Bedrock)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"aws-bedrock/claude-opus-4-5-20251101":{"id":"aws-bedrock/claude-opus-4-5-20251101","name":"Claude Opus 4.5 (AWS Bedrock)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":31999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2025-11-01","last_updated":"2025-11-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-sonnet-4-6":{"id":"anthropic/claude-sonnet-4-6","name":"Claude Sonnet 4.6 (Anthropic)","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"anthropic/claude-opus-5":{"id":"anthropic/claude-opus-5","name":"Claude Opus 5 (Anthropic)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-fable-5-1":{"id":"anthropic/claude-fable-5-1","name":"Claude Fable 5.1 (Anthropic)","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5}},"anthropic/claude-opus-4-6":{"id":"anthropic/claude-opus-4-6","name":"Claude Opus 4.6 (Anthropic)","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-sonnet-4-5-20250929":{"id":"anthropic/claude-sonnet-4-5-20250929","name":"Claude Sonnet 4.5 (2025-09-29) (Anthropic)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"anthropic/claude-opus-4-7":{"id":"anthropic/claude-opus-4-7","name":"Claude Opus 4.7 (Anthropic)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-haiku-4-5-20251001":{"id":"anthropic/claude-haiku-4-5-20251001","name":"Claude Haiku 4.5 (2025-10-01) (Anthropic)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"anthropic/claude-fable-5":{"id":"anthropic/claude-fable-5","name":"Claude Fable 5 (Anthropic)","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"anthropic/claude-haiku-4-5":{"id":"anthropic/claude-haiku-4-5","name":"Claude Haiku 4.5 (Anthropic)","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"anthropic/claude-sonnet-4-5":{"id":"anthropic/claude-sonnet-4-5","name":"Claude Sonnet 4.5 (Anthropic)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":63999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"anthropic/claude-opus-4-8":{"id":"anthropic/claude-opus-4-8","name":"Claude Opus 4.8 (Anthropic)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-sonnet-5":{"id":"anthropic/claude-sonnet-5","name":"Claude Sonnet 5 (Anthropic)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"anthropic/claude-opus-4-5-20251101":{"id":"anthropic/claude-opus-4-5-20251101","name":"Claude Opus 4.5 (Anthropic)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":31999},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2025-11-01","last_updated":"2025-11-01","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"canopywave/kimi-k2.6":{"id":"canopywave/kimi-k2.6","name":"Kimi K2.6 (CanopyWave)","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.95,"output":4,"cache_read":0.16}},"canopywave/glm-5.2":{"id":"canopywave/glm-5.2","name":"GLM-5.2 (CanopyWave)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":32768},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"canopywave/deepseek-v4-flash":{"id":"canopywave/deepseek-v4-flash","name":"DeepSeek V4 Flash (CanopyWave)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":393216},"cost":{"input":0.14,"output":0.28,"cache_read":0.03}},"canopywave/kimi-k3":{"id":"canopywave/kimi-k3","name":"Kimi K3 (CanopyWave)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":3,"output":15,"cache_read":0.3}},"canopywave/deepseek-v4-pro":{"id":"canopywave/deepseek-v4-pro","name":"DeepSeek V4 Pro (CanopyWave)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":393216},"cost":{"input":1.74,"output":3.48,"cache_read":0.01}},"together-ai/glm-4.7":{"id":"together-ai/glm-4.7","name":"GLM-4.7 (Together AI)","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":false,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":128000},"cost":{"input":0.45,"output":2}},"together-ai/minimax-m3":{"id":"together-ai/minimax-m3","name":"MiniMax M3 (Together AI)","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":false,"structured_output":true,"temperature":true,"release_date":"2026-06-01","last_updated":"2026-06-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":524288,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"together-ai/deepseek-v4-flash":{"id":"together-ai/deepseek-v4-flash","name":"DeepSeek V4 Flash (Together AI)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"output":163840},"cost":{"input":0.14,"output":0.28,"cache_read":0.03}},"together-ai/gpt-oss-20b":{"id":"together-ai/gpt-oss-20b","name":"GPT OSS 20B (Together AI)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.05,"output":0.2}},"together-ai/kimi-k3":{"id":"together-ai/kimi-k3","name":"Kimi K3 (Together AI)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":1000000},"cost":{"input":3,"output":15,"cache_read":0.3}},"together-ai/gemma-4-31b-it":{"id":"together-ai/gemma-4-31b-it","name":"Gemma 4 31B IT (Together AI)","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.39,"output":0.97}},"together-ai/deepseek-v4-pro":{"id":"together-ai/deepseek-v4-pro","name":"DeepSeek V4 Pro (Together AI)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":163840},"cost":{"input":1.32,"output":3.96,"cache_read":0.13}},"together-ai/gpt-oss-120b":{"id":"together-ai/gpt-oss-120b","name":"GPT OSS 120B (Together AI)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":0.6}},"meta/muse-spark-1.3":{"id":"meta/muse-spark-1.3","name":"Muse Spark 1.3 (Meta)","description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":1048576},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"meta/muse-spark-1.2":{"id":"meta/muse-spark-1.2","name":"Muse Spark 1.2 (Meta)","description":"Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-05","last_updated":"2026-08-05","modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"meta/muse-spark-1.1":{"id":"meta/muse-spark-1.1","name":"Muse Spark 1.1 (Meta)","description":"Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-08","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"google-ai-studio/gemini-pro-latest":{"id":"google-ai-studio/gemini-pro-latest","name":"Gemini Pro Latest (Google AI Studio)","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-27","last_updated":"2026-02-27","modalities":{"input":["text","image","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2}},"google-ai-studio/gemini-3.1-pro-preview":{"id":"google-ai-studio/gemini-3.1-pro-preview","name":"Gemini 3.1 Pro (Preview) (Google AI Studio)","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"google-ai-studio/gemini-2.5-flash-lite":{"id":"google-ai-studio/gemini-2.5-flash-lite","name":"Gemini 2.5 Flash Lite (Google AI Studio)","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65535},"cost":{"input":0.1,"output":0.4,"cache_read":0.01}},"google-ai-studio/gemini-3.6-flash":{"id":"google-ai-studio/gemini-3.6-flash","name":"Gemini 3.6 Flash (Google AI Studio)","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.08333}},"google-ai-studio/gemini-3.1-flash-lite":{"id":"google-ai-studio/gemini-3.1-flash-lite","name":"Gemini 3.1 Flash Lite (Google AI Studio)","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"cache_write":0.08333}},"google-ai-studio/gemini-3.5-flash":{"id":"google-ai-studio/gemini-3.5-flash","name":"Gemini 3.5 Flash (Google AI Studio)","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"cache_write":0.08333}},"google-ai-studio/gemini-3.5-flash-lite":{"id":"google-ai-studio/gemini-3.5-flash-lite","name":"Gemini 3.5 Flash Lite (Google AI Studio)","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"cache_write":0.08333}},"google-ai-studio/gemini-3-flash-preview":{"id":"google-ai-studio/gemini-3-flash-preview","name":"Gemini 3 Flash (Preview) (Google AI Studio)","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65535},"cost":{"input":0.5,"output":3,"cache_read":0.05}},"google-ai-studio/gemini-3.8-flash":{"id":"google-ai-studio/gemini-3.8-flash","name":"Gemini 3.8 Flash (Google AI Studio)","description":"Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.08333}},"google-ai-studio/gemini-3.7-flash":{"id":"google-ai-studio/gemini-3.7-flash","name":"Gemini 3.7 Flash (Google AI Studio)","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"cache_write":0.08333}},"google-ai-studio/gemini-2.5-pro":{"id":"google-ai-studio/gemini-2.5-pro","name":"Gemini 2.5 Pro (Google AI Studio)","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]},{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}}},"google-ai-studio/gemini-2.5-flash":{"id":"google-ai-studio/gemini-2.5-flash","name":"Gemini 2.5 Flash (Google AI Studio)","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1,"max":24576},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65535},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"bytedance/glm-4.7":{"id":"bytedance/glm-4.7","name":"GLM-4.7 (ByteDance)","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"bytedance/seed-1-8-251228":{"id":"bytedance/seed-1-8-251228","name":"Seed 1.8 (251228) (ByteDance)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-18","last_updated":"2025-12-18","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.25,"output":2,"cache_read":0.05}},"bytedance/glm-5.2":{"id":"bytedance/glm-5.2","name":"GLM-5.2 (ByteDance)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1024000,"output":128000},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"bytedance/deepseek-v4-flash":{"id":"bytedance/deepseek-v4-flash","name":"DeepSeek V4 Flash (ByteDance)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":393216},"cost":{"input":0.44,"output":1.32,"cache_read":0.014}},"bytedance/seed-1-6-flash-250715":{"id":"bytedance/seed-1-6-flash-250715","name":"Seed 1.6 Flash (250715) (ByteDance)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-26","last_updated":"2025-07-26","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.07,"output":0.3,"cache_read":0.015}},"bytedance/deepseek-v3.2":{"id":"bytedance/deepseek-v3.2","name":"DeepSeek V3.2 (ByteDance)","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.28,"output":0.42,"cache_read":0.056}},"bytedance/seed-1-6-250615":{"id":"bytedance/seed-1-6-250615","name":"Seed 1.6 (250615) (ByteDance)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-06-25","last_updated":"2025-06-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.25,"output":2,"cache_read":0.05}},"bytedance/deepseek-v4-pro":{"id":"bytedance/deepseek-v4-pro","name":"DeepSeek V4 Pro (ByteDance)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":393216},"cost":{"input":1.32,"output":3.96,"cache_read":0.044}},"bytedance/gpt-oss-120b":{"id":"bytedance/gpt-oss-120b","name":"GPT OSS 120B (ByteDance)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":32000},"cost":{"input":0.1,"output":0.5,"cache_read":0.02}},"bytedance/seed-1-6-250915":{"id":"bytedance/seed-1-6-250915","name":"Seed 1.6 (250915) (ByteDance)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-09-15","last_updated":"2025-09-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.25,"output":2,"cache_read":0.05}},"novita/glm-4.7":{"id":"novita/glm-4.7","name":"GLM-4.7 (NovitaAI)","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":128000},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"novita/qwen3.7-max":{"id":"novita/qwen3.7-max","name":"Qwen3.7 Max (NovitaAI)","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":1.25,"output":3.75,"cache_read":0.25}},"novita/gemma-4-26b-a4b-it":{"id":"novita/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B IT (NovitaAI)","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.13,"output":0.4}},"novita/llama-4-scout-17b-instruct":{"id":"novita/llama-4-scout-17b-instruct","name":"Llama 4 Scout 17B Instruct (NovitaAI)","description":"Open Llama with long-context vision for efficient multimodal agents","family":"llama","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.18,"output":0.59}},"novita/qwen35-397b-a17b":{"id":"novita/qwen35-397b-a17b","name":"Qwen3.5 397B A17B (NovitaAI)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"qwen3.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-16","last_updated":"2026-02-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":64000},"cost":{"input":0.6,"output":3.6}},"novita/qwen3-235b-a22b-thinking-2507":{"id":"novita/qwen3-235b-a22b-thinking-2507","name":"Qwen3 235B A22B Thinking 2507 (NovitaAI)","description":"Tool-capable chat model for instruction following and agentic application workflows","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-25","last_updated":"2025-07-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":32768},"cost":{"input":0.3,"output":3}},"novita/glm-4.6":{"id":"novita/glm-4.6","name":"GLM-4.6 (NovitaAI)","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.55,"output":2.2,"cache_read":0.11}},"novita/minimax-m2.1":{"id":"novita/minimax-m2.1","name":"MiniMax M2.1 (NovitaAI)","description":"Earlier MiniMax agent model for practical coding and productivity tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-23","last_updated":"2025-12-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03}},"novita/glm-4.6v":{"id":"novita/glm-4.6v","name":"GLM-4.6V (NovitaAI)","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-08","last_updated":"2025-12-08","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":16000},"cost":{"input":0.3,"output":0.9,"cache_read":0.055}},"novita/qwen3-next-80b-a3b-instruct":{"id":"novita/qwen3-next-80b-a3b-instruct","name":"Qwen3 Next 80B A3B Instruct (NovitaAI)","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":1.5}},"novita/ling-3.0-flash":{"id":"novita/ling-3.0-flash","name":"InclusionAI Ling 3.0 Flash (NovitaAI)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-02","last_updated":"2026-08-02","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":32768},"cost":{"input":0.06,"output":0.18,"cache_read":0.012}},"novita/qwen3.8-27b":{"id":"novita/qwen3.8-27b","name":"Qwen3.8 27B (NovitaAI)","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0.42,"output":3,"cache_read":0.085}},"novita/qwen3-235b-a22b-fp8":{"id":"novita/qwen3-235b-a22b-fp8","name":"Qwen3 235B A22B FP8 (NovitaAI)","description":"General-purpose chat model for instruction following, writing, and analysis","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-04-28","last_updated":"2025-04-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":40960,"output":20000},"cost":{"input":0.2,"output":0.8}},"novita/minimax-m2.7":{"id":"novita/minimax-m2.7","name":"MiniMax M2.7 (NovitaAI)","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131100},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"novita/kimi-k2.6":{"id":"novita/kimi-k2.6","name":"Kimi K2.6 (NovitaAI)","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.8,"output":3.4,"cache_read":0.16}},"novita/glm-5.2":{"id":"novita/glm-5.2","name":"GLM-5.2 (NovitaAI)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"novita/minimax-m2.5":{"id":"novita/minimax-m2.5","name":"MiniMax M2.5 (NovitaAI)","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131100},"cost":{"input":0.3,"output":1.2,"cache_read":0.03}},"novita/deepseek-v4-flash":{"id":"novita/deepseek-v4-flash","name":"DeepSeek V4 Flash (NovitaAI)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1050000,"output":393216},"cost":{"input":0.14,"output":0.28,"cache_read":0.028}},"novita/kimi-k2.7-code":{"id":"novita/kimi-k2.7-code","name":"Kimi K2.7 Code (NovitaAI)","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19}},"novita/llama-3.2-3b-instruct":{"id":"novita/llama-3.2-3b-instruct","name":"Llama 3.2 3B Instruct (NovitaAI)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2024-09-18","last_updated":"2024-09-18","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"output":32000},"cost":{"input":0.03,"output":0.05}},"novita/hy3":{"id":"novita/hy3","name":"Hy3 (NovitaAI)","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-06","last_updated":"2026-07-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":192000,"output":262144},"cost":{"input":0.14,"output":0.58,"cache_read":0.035}},"novita/qwen3-coder-30b-a3b-instruct":{"id":"novita/qwen3-coder-30b-a3b-instruct","name":"Qwen3 Coder 30B A3B Instruct (NovitaAI)","description":"Smaller Qwen coder for efficient local agents and repo-level fixes","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":160000,"output":32768},"cost":{"input":0.07,"output":0.27}},"novita/ernie-4.5-vl-424b-a47b":{"id":"novita/ernie-4.5-vl-424b-a47b","name":"ERNIE 4.5 VL 424B A47B (NovitaAI)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"ernie","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-06-30","last_updated":"2025-06-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":123000,"output":16000},"cost":{"input":0.42,"output":1.25}},"novita/kimi-k3":{"id":"novita/kimi-k3","name":"Kimi K3 (NovitaAI)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":3,"output":15,"cache_read":0.3}},"novita/deepseek-v3.2":{"id":"novita/deepseek-v3.2","name":"DeepSeek V3.2 (NovitaAI)","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":163840,"output":65536},"cost":{"input":0.269,"output":0.4,"cache_read":0.1345}},"novita/qwen3.6-35b-a3b":{"id":"novita/qwen3.6-35b-a3b","name":"Qwen3.6 35B A3B (NovitaAI)","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":64000},"cost":{"input":0.248,"output":1.485}},"novita/qwen3-max":{"id":"novita/qwen3-max","name":"Qwen3 Max (NovitaAI)","description":"Flagship Qwen3 model for coding agents, complex reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":65536},"cost":{"input":0.845,"output":3.38}},"novita/glm-5.3-flash":{"id":"novita/glm-5.3-flash","name":"GLM-5.3 Flash (NovitaAI)","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":0.075,"output":0.25,"cache_read":0.015}},"novita/qwen3-vl-30b-a3b-instruct":{"id":"novita/qwen3-vl-30b-a3b-instruct","name":"Qwen3 VL 30B A3B Instruct (NovitaAI)","description":"Multimodal model for analyzing text, images, documents, and rich media","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-10-05","last_updated":"2025-10-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":32768},"cost":{"input":0.2,"output":0.7}},"novita/llama-4-maverick-17b-instruct":{"id":"novita/llama-4-maverick-17b-instruct","name":"Llama 4 Maverick 17B Instruct (NovitaAI)","description":"Open multimodal Llama for strong reasoning with efficient everyday serving","family":"llama","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":8192},"cost":{"input":0.27,"output":0.85}},"novita/glm-4.5v":{"id":"novita/glm-4.5v","name":"GLM-4.5V (NovitaAI)","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-08-11","last_updated":"2025-08-11","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"output":16000},"cost":{"input":0.6,"output":1.8,"cache_read":0.11}},"novita/qwen3.8-flash":{"id":"novita/qwen3.8-flash","name":"Qwen3.8 Flash (NovitaAI)","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.47,"cache_read":0.016}},"novita/kimi-k2":{"id":"novita/kimi-k2","name":"Kimi K2 (NovitaAI)","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-11","last_updated":"2025-07-11","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":131072},"cost":{"input":0.57,"output":2.3}},"novita/gemma-4-31b-it":{"id":"novita/gemma-4-31b-it","name":"Gemma 4 31B IT (NovitaAI)","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.14,"output":0.4}},"novita/glm-5":{"id":"novita/glm-5","name":"GLM-5 (NovitaAI)","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202800,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"novita/qwen3.8-max":{"id":"novita/qwen3.8-max","name":"Qwen3.8 Max (NovitaAI)","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25}},"novita/glm-5.1":{"id":"novita/glm-5.1","name":"GLM-5.1 (NovitaAI)","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":1.38,"output":4.4,"cache_read":0.26}},"novita/qwen3-vl-235b-a22b-thinking":{"id":"novita/qwen3-vl-235b-a22b-thinking","name":"Qwen3 VL 235B A22B Thinking (NovitaAI)","description":"Qwen vision-language thinking model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.98,"output":3.95}},"novita/qwen3-vl-235b-a22b-instruct":{"id":"novita/qwen3-vl-235b-a22b-instruct","name":"Qwen3 VL 235B A22B Instruct (NovitaAI)","description":"Qwen vision-language instruct model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.3,"output":1.5}},"novita/qwen3-235b-a22b-instruct-2507":{"id":"novita/qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507 (NovitaAI)","description":"Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-21","last_updated":"2025-07-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":16384},"cost":{"input":0.09,"output":0.58}},"novita/qwen3-coder-480b-a35b-instruct":{"id":"novita/qwen3-coder-480b-a35b-instruct","name":"Qwen3 Coder 480B A35B Instruct (NovitaAI)","description":"Open Qwen coding heavyweight for repository reasoning and agentic engineering","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.38,"output":1.55}},"novita/glm-5.3":{"id":"novita/glm-5.3","name":"GLM-5.3 (NovitaAI)","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"novita/llama-3.3-70b-instruct":{"id":"novita/llama-3.3-70b-instruct","name":"Llama 3.3 70B Instruct (NovitaAI)","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":120000},"cost":{"input":0.135,"output":0.4}},"novita/llama-3-70b-instruct":{"id":"novita/llama-3-70b-instruct","name":"Llama 3 70B Instruct (NovitaAI)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2024-04-18","last_updated":"2024-04-18","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"output":8000},"cost":{"input":0.51,"output":0.74}},"novita/mimo-v2.5":{"id":"novita/mimo-v2.5","name":"MiMo V2.5 (NovitaAI)","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.168,"output":0.336,"cache_read":0.0034,"tiers":[{"input":0.8,"output":4,"cache_read":0.16,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.8,"output":4,"cache_read":0.16}}},"novita/mimo-v2.5-pro":{"id":"novita/mimo-v2.5-pro","name":"MiMo V2.5 Pro (NovitaAI)","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.522,"output":1.044,"cache_read":0.0043,"tiers":[{"input":2,"output":6,"cache_read":0.4,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2,"output":6,"cache_read":0.4}}},"ranoai/deepseek-v4-flash":{"id":"ranoai/deepseek-v4-flash","name":"DeepSeek V4 Flash (RanoAI)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":393216},"cost":{"input":0.14,"output":0.28,"cache_read":0.028}},"inference.net/llama-3.2-11b-instruct":{"id":"inference.net/llama-3.2-11b-instruct","name":"Llama 3.2 11B Instruct (Inference.net)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2024-09-25","last_updated":"2024-09-25","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":128000},"cost":{"input":0.07,"output":0.33}},"sakana/fugu-ultra":{"id":"sakana/fugu-ultra","name":"Fugu Ultra (Sakana AI)","description":"Quality-first multi-agent model for hard research, analysis, and competitions","family":"fugu","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"release_date":"2026-06-15","last_updated":"2026-06-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":1000000},"cost":{"input":5,"output":30,"cache_read":0.5}},"deepinfra/gemma-4-26b-a4b-it":{"id":"deepinfra/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B IT (DeepInfra)","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.07,"output":0.34}},"deepinfra/qwen3.5-9b":{"id":"deepinfra/qwen3.5-9b","name":"Qwen3.5 9B (DeepInfra)","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.1,"output":0.15}},"deepinfra/ling-3.0-flash":{"id":"deepinfra/ling-3.0-flash","name":"InclusionAI Ling 3.0 Flash (DeepInfra)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-02","last_updated":"2026-08-02","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":32768},"cost":{"input":0.06,"output":0.18,"cache_read":0.012}},"deepinfra/deepseek-v4-flash":{"id":"deepinfra/deepseek-v4-flash","name":"DeepSeek V4 Flash (DeepInfra)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":393216},"cost":{"input":0.08,"output":0.18,"cache_read":0.016}},"deepinfra/hy3":{"id":"deepinfra/hy3","name":"Hy3 (DeepInfra)","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-06","last_updated":"2026-07-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"input":192000,"output":131072},"cost":{"input":0.14,"output":0.58,"cache_read":0.035}},"deepinfra/deepseek-v3.2":{"id":"deepinfra/deepseek-v3.2","name":"DeepSeek V3.2 (DeepInfra)","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":160000,"output":65536},"cost":{"input":0.26,"output":0.38,"cache_read":0.13}},"deepinfra/nemotron-3-ultra-550b":{"id":"deepinfra/nemotron-3-ultra-550b","name":"Nemotron 3 Ultra 550B (DeepInfra)","description":"Nemotron multimodal model for visual reasoning and agentic AI workflows","family":"nemotron","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-06-01","last_updated":"2026-06-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":2.2,"cache_read":0.1}},"deepinfra/qwen3-vl-30b-a3b-instruct":{"id":"deepinfra/qwen3-vl-30b-a3b-instruct","name":"Qwen3 VL 30B A3B Instruct (DeepInfra)","description":"Multimodal model for analyzing text, images, documents, and rich media","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-10-05","last_updated":"2025-10-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":32768},"cost":{"input":0.15,"output":0.6}},"deepinfra/gemma-4-31b-it":{"id":"deepinfra/gemma-4-31b-it","name":"Gemma 4 31B IT (DeepInfra)","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.13,"output":0.38}},"deepinfra/kimi-k2.5":{"id":"deepinfra/kimi-k2.5","name":"Kimi K2.5 (DeepInfra)","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":131072},"cost":{"input":0.45,"output":2.25,"cache_read":0.07}},"deepinfra/glm-5.1":{"id":"deepinfra/glm-5.1","name":"GLM-5.1 (DeepInfra)","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":198000,"output":65536},"cost":{"input":1.05,"output":3.5,"cache_read":0.205}},"deepinfra/qwen3-vl-235b-a22b-instruct":{"id":"deepinfra/qwen3-vl-235b-a22b-instruct","name":"Qwen3 VL 235B A22B Instruct (DeepInfra)","description":"Qwen vision-language instruct model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.2,"output":0.88,"cache_read":0.11}},"deepinfra/deepseek-v4-pro":{"id":"deepinfra/deepseek-v4-pro","name":"DeepSeek V4 Pro (DeepInfra)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":64000},"cost":{"input":1.3,"output":2.6,"cache_read":0.1}},"deepinfra/mimo-v2.5":{"id":"deepinfra/mimo-v2.5","name":"MiMo V2.5 (DeepInfra)","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":16384},"cost":{"input":0.4,"output":2,"cache_read":0.08,"tiers":[{"input":0.8,"output":4,"cache_read":0.16,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.8,"output":4,"cache_read":0.16}}},"deepinfra/mimo-v2.5-pro":{"id":"deepinfra/mimo-v2.5-pro","name":"MiMo V2.5 Pro (DeepInfra)","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":false,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":16384},"cost":{"input":1,"output":3,"cache_read":0.2,"tiers":[{"input":2,"output":6,"cache_read":0.4,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2,"output":6,"cache_read":0.4}}},"azure-ai-foundry/grok-4-1-fast-non-reasoning":{"id":"azure-ai-foundry/grok-4-1-fast-non-reasoning","name":"Grok 4.1 Fast Non-Reasoning (Azure AI Foundry)","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","family":"grok","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-11-19","last_updated":"2025-11-19","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":30000},"cost":{"input":0.2,"output":0.5}},"azure-ai-foundry/grok-4-1-fast-reasoning":{"id":"azure-ai-foundry/grok-4-1-fast-reasoning","name":"Grok 4.1 Fast Reasoning (Azure AI Foundry)","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-11-19","last_updated":"2025-11-19","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":30000},"cost":{"input":0.2,"output":0.5}},"azure-ai-foundry/grok-4-3":{"id":"azure-ai-foundry/grok-4-3","name":"Grok 4.3 (Azure AI Foundry)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-30","last_updated":"2026-04-30","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":20000,"output":8192},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"moonshot/kimi-k2.7-code-highspeed":{"id":"moonshot/kimi-k2.7-code-highspeed","name":"Kimi K2.7 Code Highspeed (Moonshot AI)","description":"Lower-latency Kimi Code variant for interactive edits and coding-agent loops","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":1.9,"output":8,"cache_read":0.38}},"moonshot/kimi-k2.6":{"id":"moonshot/kimi-k2.6","name":"Kimi K2.6 (Moonshot AI)","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.16}},"moonshot/kimi-k2.7-code":{"id":"moonshot/kimi-k2.7-code","name":"Kimi K2.7 Code (Moonshot AI)","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19}},"moonshot/kimi-k3":{"id":"moonshot/kimi-k3","name":"Kimi K3 (Moonshot AI)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":1048576},"cost":{"input":3,"output":15,"cache_read":0.3}},"moonshot/kimi-k2.5":{"id":"moonshot/kimi-k2.5","name":"Kimi K2.5 (Moonshot AI)","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.6,"output":3,"cache_read":0.1}},"consensusprotocol/Qwen3.8-27B":{"id":"consensusprotocol/Qwen3.8-27B","name":"Qwen3.8 27B (Consensus Protocol)","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"output":32768},"cost":{"input":0.41,"output":2.5,"cache_read":0.08}},"consensusprotocol/deepseek-v4-flash":{"id":"consensusprotocol/deepseek-v4-flash","name":"DeepSeek V4 Flash (Consensus Protocol)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":524288,"output":393216},"cost":{"input":0.08,"output":0.16,"cache_read":0.02}},"consensusprotocol/gpt-oss-20b":{"id":"consensusprotocol/gpt-oss-20b","name":"GPT OSS 20B (Consensus Protocol)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"output":32768},"cost":{"input":0.04,"output":0.19,"cache_read":0.01}},"consensusprotocol/glm-5.3-flash":{"id":"consensusprotocol/glm-5.3-flash","name":"GLM-5.3 Flash (Consensus Protocol)","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":0.1,"output":0.25,"cache_read":0.02}},"azure/gpt-5-nano":{"id":"azure/gpt-5-nano","name":"GPT-5 Nano (Azure)","description":"Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs","family":"gpt-nano","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005}},"azure/gpt-4.1-nano":{"id":"azure/gpt-4.1-nano","name":"GPT-4.1 Nano (Azure)","description":"Tiny GPT-4.1 option for classification, routing, and very high-volume tasks","family":"gpt-nano","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.025}},"azure/gpt-5.1-codex-mini":{"id":"azure/gpt-5.1-codex-mini","name":"GPT-5.1 Codex mini (Azure)","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"azure/gpt-5.1-codex":{"id":"azure/gpt-5.1-codex","name":"GPT-5.1 Codex (Azure)","description":"Codex GPT for repository edits, code review, and practical software agents","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":272000},"cost":{"input":1.25,"output":10}},"azure/gpt-5.6-sol":{"id":"azure/gpt-5.6-sol","name":"GPT-5.6 Sol (Azure)","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"cache_write":6.25}},"azure/gpt-5.2-codex":{"id":"azure/gpt-5.2-codex","name":"GPT-5.2 Codex (Azure)","description":"Code-specialist GPT for repository edits, reviews, and long-running software agents","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"azure/gpt-5.2-pro":{"id":"azure/gpt-5.2-pro","name":"GPT-5.2 Pro (Azure)","description":"Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":272000},"cost":{"input":21,"output":168}},"azure/gpt-4.1-mini":{"id":"azure/gpt-4.1-mini","name":"GPT-4.1 Mini (Azure)","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1}},"azure/gpt-5.4":{"id":"azure/gpt-5.4","name":"GPT-5.4 (Azure)","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25}},"azure/gpt-4-turbo":{"id":"azure/gpt-4-turbo","name":"GPT-4 Turbo (Azure)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-12","release_date":"2023-11-06","last_updated":"2024-04-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096},"cost":{"input":10,"output":30}},"azure/gpt-5.1":{"id":"azure/gpt-5.1","name":"GPT-5.1 (Azure)","description":"Sharper GPT-5 generation for coding, product work, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"azure/o1":{"id":"azure/o1","name":"o1 (Azure)","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":true,"temperature":false,"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5}},"azure/gpt-4o":{"id":"azure/gpt-4o","name":"GPT-4o (Azure)","description":"Omni-era GPT for multimodal chat, practical coding, and general assistants","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"azure/gpt-5.6-luna":{"id":"azure/gpt-5.6-luna","name":"GPT-5.6 Luna (Azure)","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25}},"azure/gpt-5.3-codex":{"id":"azure/gpt-5.3-codex","name":"GPT-5.3 Codex (Azure)","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"azure/gpt-4.1":{"id":"azure/gpt-4.1","name":"GPT-4.1 (Azure)","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5}},"azure/gpt-5.4-nano":{"id":"azure/gpt-5.4-nano","name":"GPT-5.4 Nano (Azure)","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02}},"azure/gpt-5.4-mini":{"id":"azure/gpt-5.4-mini","name":"GPT-5.4 Mini (Azure)","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075}},"azure/gpt-3.5-turbo":{"id":"azure/gpt-3.5-turbo","name":"GPT-3.5 Turbo (Azure)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2021-09-01","release_date":"2023-03-01","last_updated":"2023-11-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":16385,"output":4096},"cost":{"input":0.5,"output":1.5}},"azure/gpt-5-mini":{"id":"azure/gpt-5-mini","name":"GPT-5 Mini (Azure)","description":"Small GPT-5 for responsive agents, coding help, and everyday automation","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"azure/gpt-oss-120b":{"id":"azure/gpt-oss-120b","name":"GPT OSS 120B (Azure)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":0.6}},"azure/gpt-5.4-pro":{"id":"azure/gpt-5.4-pro","name":"GPT-5.4 Pro (Azure)","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180}},"azure/gpt-5.6-terra":{"id":"azure/gpt-5.6-terra","name":"GPT-5.6 Terra (Azure)","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5}},"azure/gpt-4":{"id":"azure/gpt-4","name":"GPT-4 (Azure)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-11","release_date":"2023-11-06","last_updated":"2024-04-09","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"output":8192},"cost":{"input":30,"output":60}},"azure/gpt-5.2":{"id":"azure/gpt-5.2","name":"GPT-5.2 (Azure)","description":"Reliable GPT generation for broad coding, writing, and tool-assisted product work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"azure/gpt-5":{"id":"azure/gpt-5","name":"GPT-5 (Azure)","description":"Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"azure/o4-mini":{"id":"azure/o4-mini","name":"o4 Mini (Azure)","description":"Fast o-series model for compact reasoning, coding, and tool use","family":"o-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.275}},"azure/o3-mini":{"id":"azure/o3-mini","name":"o3 Mini (Azure)","description":"Smaller o-series reasoner for economical coding, math, and planning tasks","family":"o-mini","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"azure/o3":{"id":"azure/o3","name":"o3 (Azure)","description":"Deliberate o-series reasoner for hard math, coding, and multi-step analysis","family":"o","attachment":true,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":0.5}},"azure/gpt-5.5":{"id":"azure/gpt-5.5","name":"GPT-5.5 (Azure)","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}}},"deepseek/deepseek-v4-flash-vision-exp":{"id":"deepseek/deepseek-v4-flash-vision-exp","name":"DeepSeek V4 Flash Vision Exp (DeepSeek)","description":"Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work","family":"deepseek-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-21","last_updated":"2026-08-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"output":393216},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028}},"deepseek/deepseek-v4-flash":{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek V4 Flash (DeepSeek)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1050000,"output":393216},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028}},"deepseek/deepseek-v4-pro":{"id":"deepseek/deepseek-v4-pro","name":"DeepSeek V4 Pro (DeepSeek)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1050000,"output":393216},"cost":{"input":0.435,"output":0.87,"cache_read":0.003625}},"openai/gpt-5-nano":{"id":"openai/gpt-5-nano","name":"GPT-5 Nano (OpenAI)","description":"Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs","family":"gpt-nano","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005}},"openai/gpt-4.1-nano":{"id":"openai/gpt-4.1-nano","name":"GPT-4.1 Nano (OpenAI)","description":"Tiny GPT-4.1 option for classification, routing, and very high-volume tasks","family":"gpt-nano","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.025}},"openai/gpt-5-pro":{"id":"openai/gpt-5-pro","name":"GPT-5 Pro (OpenAI)","description":"Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-10-06","last_updated":"2025-10-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":272000},"cost":{"input":15,"output":120}},"openai/gpt-4o-mini-transcribe":{"id":"openai/gpt-4o-mini-transcribe","name":"GPT-4o Mini Transcribe (OpenAI)","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"gpt","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-03-20","last_updated":"2025-03-20","modalities":{"input":["text","audio"],"output":["text"]},"open_weights":false,"limit":{"context":16000,"output":2000},"cost":{"input":1.25,"output":5}},"openai/gpt-5.6-sol":{"id":"openai/gpt-5.6-sol","name":"GPT-5.6 Sol (OpenAI)","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"cache_write":6.25}},"openai/gpt-6-astra":{"id":"openai/gpt-6-astra","name":"GPT-6 Astra (OpenAI)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-09-03","last_updated":"2026-09-03","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"openai/gpt-5.2-pro":{"id":"openai/gpt-5.2-pro","name":"GPT-5.2 Pro (OpenAI)","description":"Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":272000},"cost":{"input":21,"output":168}},"openai/gpt-4.1-mini":{"id":"openai/gpt-4.1-mini","name":"GPT-4.1 Mini (OpenAI)","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1}},"openai/gpt-5.4":{"id":"openai/gpt-5.4","name":"GPT-5.4 (OpenAI)","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25}},"openai/gpt-4o-transcribe":{"id":"openai/gpt-4o-transcribe","name":"GPT-4o Transcribe (OpenAI)","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"gpt","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-03-20","last_updated":"2025-03-20","modalities":{"input":["text","audio"],"output":["text"]},"open_weights":false,"limit":{"context":16000,"output":2000},"cost":{"input":2.5,"output":10}},"openai/gpt-4-turbo":{"id":"openai/gpt-4-turbo","name":"GPT-4 Turbo (OpenAI)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-12","release_date":"2023-11-06","last_updated":"2024-04-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096},"cost":{"input":10,"output":30}},"openai/gpt-5.1":{"id":"openai/gpt-5.1","name":"GPT-5.1 (OpenAI)","description":"Sharper GPT-5 generation for coding, product work, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/o1":{"id":"openai/o1","name":"o1 (OpenAI)","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":false,"structured_output":true,"temperature":false,"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5}},"openai/gpt-4o":{"id":"openai/gpt-4o","name":"GPT-4o (OpenAI)","description":"Omni-era GPT for multimodal chat, practical coding, and general assistants","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25}},"openai/gpt-5.6-luna":{"id":"openai/gpt-5.6-luna","name":"GPT-5.6 Luna (OpenAI)","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25}},"openai/gpt-5.3-codex":{"id":"openai/gpt-5.3-codex","name":"GPT-5.3 Codex (OpenAI)","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-4o-mini":{"id":"openai/gpt-4o-mini","name":"GPT-4o Mini (OpenAI)","description":"Small omni GPT for cheap multimodal assistance and production-scale traffic","family":"gpt-mini","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"openai/gpt-4.1":{"id":"openai/gpt-4.1","name":"GPT-4.1 (OpenAI)","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5}},"openai/gpt-5.4-nano":{"id":"openai/gpt-5.4-nano","name":"GPT-5.4 Nano (OpenAI)","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02}},"openai/gpt-5.5-pro":{"id":"openai/gpt-5.5-pro","name":"GPT-5.5 Pro (OpenAI)","description":"Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}}},"openai/gpt-5.4-mini":{"id":"openai/gpt-5.4-mini","name":"GPT-5.4 Mini (OpenAI)","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075}},"openai/gpt-3.5-turbo":{"id":"openai/gpt-3.5-turbo","name":"GPT-3.5 Turbo (OpenAI)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2021-09-01","release_date":"2023-03-01","last_updated":"2023-11-06","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":16385,"output":4096},"cost":{"input":0.5,"output":1.5}},"openai/gpt-5-mini":{"id":"openai/gpt-5-mini","name":"GPT-5 Mini (OpenAI)","description":"Small GPT-5 for responsive agents, coding help, and everyday automation","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"openai/gpt-5.4-pro":{"id":"openai/gpt-5.4-pro","name":"GPT-5.4 Pro (OpenAI)","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180}},"openai/gpt-5.6-terra":{"id":"openai/gpt-5.6-terra","name":"GPT-5.6 Terra (OpenAI)","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5}},"openai/gpt-4":{"id":"openai/gpt-4","name":"GPT-4 (OpenAI)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-11","release_date":"2023-11-06","last_updated":"2024-04-09","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":8192,"output":8192},"cost":{"input":30,"output":60}},"openai/gpt-5.2":{"id":"openai/gpt-5.2","name":"GPT-5.2 (OpenAI)","description":"Reliable GPT generation for broad coding, writing, and tool-assisted product work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"openai/gpt-5":{"id":"openai/gpt-5","name":"GPT-5 (OpenAI)","description":"Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"openai/o4-mini":{"id":"openai/o4-mini","name":"o4 Mini (OpenAI)","description":"Fast o-series model for compact reasoning, coding, and tool use","family":"o-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.275}},"openai/o3-mini":{"id":"openai/o3-mini","name":"o3 Mini (OpenAI)","description":"Smaller o-series reasoner for economical coding, math, and planning tasks","family":"o-mini","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55}},"openai/o3":{"id":"openai/o3","name":"o3 (OpenAI)","description":"Deliberate o-series reasoner for hard math, coding, and multi-step analysis","family":"o","attachment":true,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":false,"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":0.5}},"openai/gpt-5.5":{"id":"openai/gpt-5.5","name":"GPT-5.5 (OpenAI)","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}}},"meta-contributor/muse-spark-1.2-contributor":{"id":"meta-contributor/muse-spark-1.2-contributor","name":"Muse Spark 1.2 Contributor (Meta Contributor)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-06","last_updated":"2026-08-06","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":0.1,"output":0.2,"cache_read":0.002}},"meta-contributor/muse-spark-1.3-contributor":{"id":"meta-contributor/muse-spark-1.3-contributor","name":"Muse Spark 1.3 Contributor (Meta Contributor)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-09-02","last_updated":"2026-09-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":1048576},"cost":{"input":0.1,"output":0.2,"cache_read":0.002}},"xai/grok-4":{"id":"xai/grok-4","name":"Grok 4 (xAI)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-09","last_updated":"2025-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":3,"output":15,"cache_read":0.75}},"xai/grok-4-5":{"id":"xai/grok-4-5","name":"Grok 4.5 (xAI)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-07-08","last_updated":"2026-07-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.3}},"xai/grok-build-0-1":{"id":"xai/grok-build-0-1","name":"Grok Build 0.1 (xAI)","description":"Grok coding model for agentic engineering, edits, and codebase workflows","family":"grok-build","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-05-20","last_updated":"2026-05-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":1,"output":2,"cache_read":0.2,"tiers":[{"input":2,"output":4,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2,"output":4,"cache_read":0.4}}},"xai/grok-4-3":{"id":"xai/grok-4-3","name":"Grok 4.3 (xAI)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-30","last_updated":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":1000000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"xai/grok-4-20-beta-0309-reasoning":{"id":"xai/grok-4-20-beta-0309-reasoning","name":"Grok 4.20 Beta Reasoning (0309) (xAI)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-09","last_updated":"2026-03-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":30000},"cost":{"input":2,"output":6,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"xai/grok-4-6":{"id":"xai/grok-4-6","name":"Grok 4.6 (xAI)","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5}},"xai/grok-4-20-beta-0309-non-reasoning":{"id":"xai/grok-4-20-beta-0309-non-reasoning","name":"Grok 4.20 Beta Non-Reasoning (0309) (xAI)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-03-09","last_updated":"2026-03-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":2000000,"output":30000},"cost":{"input":2,"output":6,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"zai/glm-4.7":{"id":"zai/glm-4.7","name":"GLM-4.7 (Z AI)","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"zai/glm-4.5-air":{"id":"zai/glm-4.5-air","name":"GLM-4.5 Air (Z AI)","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":98304},"cost":{"input":0.2,"output":1.1,"cache_read":0.03}},"zai/glm-4.6":{"id":"zai/glm-4.6","name":"GLM-4.6 (Z AI)","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"zai/glm-4.6v":{"id":"zai/glm-4.6v","name":"GLM-4.6V (Z AI)","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-08","last_updated":"2025-12-08","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":16000},"cost":{"input":0.3,"output":0.9,"cache_read":0.05}},"zai/glm-4.6v-flashx":{"id":"zai/glm-4.6v-flashx","name":"GLM-4.6V FlashX (Z AI)","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-12-08","last_updated":"2025-12-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16000},"cost":{"input":0.04,"output":0.4,"cache_read":0.004}},"zai/glm-5.2":{"id":"zai/glm-5.2","name":"GLM-5.2 (Z AI)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":128000},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"zai/glm-4.5-x":{"id":"zai/glm-4.5-x","name":"GLM-4.5 X (Z AI)","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":128000},"cost":{"input":2.2,"output":8.9,"cache_read":0.45}},"zai/glm-4.5-airx":{"id":"zai/glm-4.5-airx","name":"GLM-4.5 AirX (Z AI)","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":128000},"cost":{"input":1.1,"output":4.5,"cache_read":0.22}},"zai/glm-5.3-flash":{"id":"zai/glm-5.3-flash","name":"GLM-5.3 Flash (Z AI)","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03}},"zai/glm-4.5":{"id":"zai/glm-4.5","name":"GLM-4.5 (Z AI)","description":"Hybrid-reasoning GLM release that made the 4.5 line broadly useful","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":98304},"cost":{"input":0.6,"output":2.2,"cache_read":0.11}},"zai/glm-4.5v":{"id":"zai/glm-4.5v","name":"GLM-4.5V (Z AI)","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-08-11","last_updated":"2025-08-11","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":16000},"cost":{"input":0.6,"output":1.8,"cache_read":0.11}},"zai/glm-4.7-flashx":{"id":"zai/glm-4.7-flashx","name":"GLM-4.7 FlashX (Z AI)","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":0.07,"output":0.4,"cache_read":0.01}},"zai/glm-5":{"id":"zai/glm-5","name":"GLM-5 (Z AI)","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202800,"output":131100},"cost":{"input":1,"output":3.2,"cache_read":0.2}},"zai/glm-4-32b-0414-128k":{"id":"zai/glm-4-32b-0414-128k","name":"GLM-4 32B (0414-128k) (Z AI)","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-04-14","last_updated":"2025-04-14","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":128000},"cost":{"input":0.1,"output":0.1}},"zai/glm-5.1":{"id":"zai/glm-5.1","name":"GLM-5.1 (Z AI)","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"zai/glm-5.3":{"id":"zai/glm-5.3","name":"GLM-5.3 (Z AI)","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":128000},"cost":{"input":1.4,"output":4.4,"cache_read":0.26}},"azure-anthropic/claude-opus-5":{"id":"azure-anthropic/claude-opus-5","name":"Claude Opus 5 (Azure Anthropic)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"azure-anthropic/claude-opus-4-6":{"id":"azure-anthropic/claude-opus-4-6","name":"Claude Opus 4.6 (Azure Anthropic)","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"azure-anthropic/claude-opus-4-7":{"id":"azure-anthropic/claude-opus-4-7","name":"Claude Opus 4.7 (Azure Anthropic)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"azure-anthropic/claude-fable-5":{"id":"azure-anthropic/claude-fable-5","name":"Claude Fable 5 (Azure Anthropic)","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"azure-anthropic/claude-opus-4-8":{"id":"azure-anthropic/claude-opus-4-8","name":"Claude Opus 4.8 (Azure Anthropic)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"azure-anthropic/claude-sonnet-5":{"id":"azure-anthropic/claude-sonnet-5","name":"Claude Sonnet 5 (Azure Anthropic)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"fireworks/deepseek-v4-flash":{"id":"fireworks/deepseek-v4-flash","name":"DeepSeek V4 Flash (Fireworks AI)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":393216},"cost":{"input":0.22,"output":0.66,"cache_read":0.007}},"fireworks/kimi-k3":{"id":"fireworks/kimi-k3","name":"Kimi K3 (Fireworks AI)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1040384,"output":1040384},"cost":{"input":3,"output":15,"cache_read":0.3}},"fireworks/kimi-k3-fast":{"id":"fireworks/kimi-k3-fast","name":"Kimi K3 Fast (Fireworks AI)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1040384,"output":1040384},"cost":{"input":4.5,"output":22.5,"cache_read":0.45}},"fireworks/deepseek-v4-pro":{"id":"fireworks/deepseek-v4-pro","name":"DeepSeek V4 Pro (Fireworks AI)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":393216},"cost":{"input":1.32,"output":3.96,"cache_read":0.044}},"mistral/ministral-14b-2512":{"id":"mistral/ministral-14b-2512","name":"Ministral 14B (Mistral AI)","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-12-02","last_updated":"2025-12-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":262144},"cost":{"input":0.2,"output":0.2}},"mistral/codestral-2508":{"id":"mistral/codestral-2508","name":"Codestral (Mistral AI)","description":"Mistral coding model for code completion, generation, and developer workflows","family":"codestral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-07-30","last_updated":"2025-07-30","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.3,"output":0.9}},"mistral/mistral-small-2506":{"id":"mistral/mistral-small-2506","name":"Mistral Small 3.2 (Mistral AI)","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-03","release_date":"2025-06-20","last_updated":"2025-06-20","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":16384},"cost":{"input":0.1,"output":0.3}},"mistral/devstral-2512":{"id":"mistral/devstral-2512","name":"Devstral 2 (Mistral AI)","description":"Mistral's coding-agent model for repository work, terminal tasks, and software fixes","family":"devstral","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2025-12","release_date":"2025-12-09","last_updated":"2025-12-09","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":2}},"mistral/mistral-large-2512":{"id":"mistral/mistral-large-2512","name":"Mistral Large 3 (Mistral AI)","description":"Mistral's largest general model for enterprise agents, coding, and multilingual reasoning","family":"mistral-large","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-11","release_date":"2024-11-01","last_updated":"2025-12-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":1.5}},"mistral/ministral-3b-2512":{"id":"mistral/ministral-3b-2512","name":"Ministral 3B (Mistral AI)","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-12-02","last_updated":"2025-12-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":131072},"cost":{"input":0.1,"output":0.1}},"mistral/mistral-large-latest":{"id":"mistral/mistral-large-latest","name":"Mistral Large Latest (Mistral AI)","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","attachment":false,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"knowledge":"2024-11","release_date":"2024-11-01","last_updated":"2025-12-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":262144},"cost":{"input":4,"output":12}},"mistral/ministral-8b-2512":{"id":"mistral/ministral-8b-2512","name":"Ministral 8B (Mistral AI)","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":true,"reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"release_date":"2025-12-02","last_updated":"2025-12-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":262144},"cost":{"input":0.15,"output":0.15}},"cerebras/glm-4.7":{"id":"cerebras/glm-4.7","name":"GLM-4.7 (Cerebras)","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":2.25,"output":2.75}},"cerebras/gemma-4-31b-it":{"id":"cerebras/gemma-4-31b-it","name":"Gemma 4 31B IT (Cerebras)","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.99,"output":1.49}},"cerebras/qwen3-235b-a22b-instruct-2507":{"id":"cerebras/qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507 (Cerebras)","description":"Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-07-21","last_updated":"2025-07-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262000,"output":8192},"cost":{"input":0.6,"output":1.2}},"cerebras/gpt-oss-120b":{"id":"cerebras/gpt-oss-120b","name":"GPT OSS 120B (Cerebras)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.35,"output":0.75}},"cerebras/llama-3.3-70b-instruct":{"id":"cerebras/llama-3.3-70b-instruct","name":"Llama 3.3 70B Instruct (Cerebras)","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0.85,"output":1.2}},"perplexity/sonar":{"id":"perplexity/sonar","name":"Sonar (Perplexity)","description":"Fast web-grounded Sonar for current answers, citations, and lightweight retrieval","family":"sonar","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":130000,"output":4096},"cost":{"input":1,"output":1}},"perplexity/sonar-reasoning-pro":{"id":"perplexity/sonar-reasoning-pro","name":"Sonar Reasoning Pro (Perplexity)","description":"Web-grounded Sonar for multi-step research questions that need cited reasoning","family":"sonar-reasoning","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":4096},"cost":{"input":2,"output":8}},"perplexity/sonar-pro":{"id":"perplexity/sonar-pro","name":"Sonar Pro (Perplexity)","description":"Deeper Sonar search model with broader retrieval and stronger synthesis","family":"sonar-pro","attachment":false,"reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":8192},"cost":{"input":3,"output":15}},"runware/kimi-k2.6":{"id":"runware/kimi-k2.6","name":"Kimi K2.6 (Runware)","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":131072},"cost":{"input":0.6,"output":3.05,"cache_read":0.13}},"runware/glm-5.2":{"id":"runware/glm-5.2","name":"GLM-5.2 (Runware)","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":false,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1024000,"output":128000},"cost":{"input":0.8,"output":2.55,"cache_read":0.16}},"runware/deepseek-v4-flash":{"id":"runware/deepseek-v4-flash","name":"DeepSeek V4 Flash (Runware)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":384000},"cost":{"input":0.076,"output":0.153,"cache_read":0.014}},"runware/glm-5.3-flash":{"id":"runware/glm-5.3-flash","name":"GLM-5.3 Flash (Runware)","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03}},"runware/gemma-4-31b-it":{"id":"runware/gemma-4-31b-it","name":"Gemma 4 31B IT (Runware)","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.102,"output":0.297,"cache_read":0.012}},"runware/deepseek-v4-pro":{"id":"runware/deepseek-v4-pro","name":"DeepSeek V4 Pro (Runware)","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":384000},"cost":{"input":0.961,"output":1.922,"cache_read":0.079}},"runware/gpt-oss-120b":{"id":"runware/gpt-oss-120b","name":"GPT OSS 120B (Runware)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-08-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.032,"output":0.14,"cache_read":0.032}},"runware/glm-5.3":{"id":"runware/glm-5.3","name":"GLM-5.3 (Runware)","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.2}}}},"llama":{"id":"llama","env":["LLAMA_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.llama.com/compat/v1/","name":"Llama","doc":"https://llama.developer.meta.com/docs/models","models":{"cerebras-llama-4-scout-17b-16e-instruct":{"id":"cerebras-llama-4-scout-17b-16e-instruct","name":"Cerebras-Llama-4-Scout-17B-16E-Instruct","description":"Open multimodal Llama model for long-context analysis and efficient agents","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0,"output":0}},"llama-4-maverick-17b-128e-instruct-fp8":{"id":"llama-4-maverick-17b-128e-instruct-fp8","name":"Llama-4-Maverick-17B-128E-Instruct-FP8","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0,"output":0}},"groq-llama-4-maverick-17b-128e-instruct":{"id":"groq-llama-4-maverick-17b-128e-instruct","name":"Groq-Llama-4-Maverick-17B-128E-Instruct","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0,"output":0}},"llama-4-scout-17b-16e-instruct-fp8":{"id":"llama-4-scout-17b-16e-instruct-fp8","name":"Llama-4-Scout-17B-16E-Instruct-FP8","description":"Open multimodal Llama model for long-context analysis and efficient agents","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0,"output":0}},"cerebras-llama-4-maverick-17b-128e-instruct":{"id":"cerebras-llama-4-maverick-17b-128e-instruct","name":"Cerebras-Llama-4-Maverick-17B-128E-Instruct","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-04-05","last_updated":"2025-04-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0,"output":0}},"llama-3.3-70b-instruct":{"id":"llama-3.3-70b-instruct","name":"Llama-3.3-70B-Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0,"output":0}},"llama-3.3-8b-instruct":{"id":"llama-3.3-8b-instruct","name":"Llama-3.3-8B-Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":true,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":128000,"output":4096},"cost":{"input":0,"output":0}}}},"alibaba-token-plan":{"id":"alibaba-token-plan","env":["ALIBABA_TOKEN_PLAN_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1","name":"Alibaba Token Plan","doc":"https://www.alibabacloud.com/help/en/model-studio/token-plan-overview","models":{"qwen3.7-max":{"id":"qwen3.7-max","name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"happyhorse-1.1-r2v":{"id":"happyhorse-1.1-r2v","name":"HappyHorse 1.1 Reference-to-Video","description":"Video model for reference-guided video generation","attachment":true,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-07-17","last_updated":"2026-07-17","modalities":{"input":["image","text"],"output":["video"]},"open_weights":false,"limit":{"context":0,"output":0},"cost":{"input":0,"output":0}},"deepseek-v4-pro-0813":{"id":"deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"deepseek-v4-flash-0731":{"id":"deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"qwen3.8-max-preview":{"id":"qwen3.8-max-preview","name":"Qwen3.8 Max Preview","description":"Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens","min":0,"max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-07-19","last_updated":"2026-07-19","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"status":"beta","cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"qwen3.6-plus":{"id":"qwen3.6-plus","name":"Qwen3.6 Plus","description":"Earlier Qwen multimodal workhorse for million-token agent and document tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":131072}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"wan2.7-image-pro":{"id":"wan2.7-image-pro","name":"Wan2.7 Image Pro","description":"Image model for prompt-driven generation, editing, and visual design workflows","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-05-29","last_updated":"2026-05-29","modalities":{"input":["text"],"output":["image"]},"open_weights":false,"limit":{"context":8192,"output":0},"cost":{"input":0,"output":0}},"kimi-k2.6":{"id":"kimi-k2.6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"happyhorse-1.1-t2v":{"id":"happyhorse-1.1-t2v","name":"HappyHorse 1.1 Text-to-Video","description":"Video model for prompt-driven text-to-video generation","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-07-17","last_updated":"2026-07-17","modalities":{"input":["text"],"output":["video"]},"open_weights":false,"limit":{"context":0,"output":0},"cost":{"input":0,"output":0}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"deepseek-v4-flash":{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"kimi-k2.7-code":{"id":"kimi-k2.7-code","name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"qwen-image-2.0-pro":{"id":"qwen-image-2.0-pro","name":"Qwen Image 2.0 Pro","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-03-03","last_updated":"2026-03-03","modalities":{"input":["text"],"output":["image"]},"open_weights":false,"limit":{"context":8192,"output":0},"cost":{"input":0,"output":0}},"deepseek-v3.2":{"id":"deepseek-v3.2","name":"DeepSeek V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-12-03","last_updated":"2025-12-05","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":65536},"cost":{"input":0,"output":0}},"MiniMax-M2.5":{"id":"MiniMax-M2.5","name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":196608,"input":196601,"output":32768},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"happyhorse-1.1-i2v":{"id":"happyhorse-1.1-i2v","name":"HappyHorse 1.1 Image-to-Video","description":"Video model for image-to-video generation","attachment":true,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-07-17","last_updated":"2026-07-17","modalities":{"input":["image","text"],"output":["video"]},"open_weights":false,"limit":{"context":0,"output":0},"cost":{"input":0,"output":0}},"qwen-image-2.0":{"id":"qwen-image-2.0","name":"Qwen Image 2.0","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"qwen","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-03-03","last_updated":"2026-03-03","modalities":{"input":["text"],"output":["image"]},"open_weights":false,"limit":{"context":8192,"output":0},"cost":{"input":0,"output":0}},"qwen3.6-flash":{"id":"qwen3.6-flash","name":"Qwen3.6 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":131072}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-27","last_updated":"2026-04-27","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"qwen3.8-flash":{"id":"qwen3.8-flash","name":"Qwen3.8 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens","max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-5":{"id":"glm-5","name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":202752,"output":16384},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"qwen3.8-max":{"id":"qwen3.8-max","name":"Qwen3.8 Max","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens","min":0,"max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"kimi-k2.5":{"id":"kimi-k2.5","name":"Kimi K2.5","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":98304},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-5.1":{"id":"glm-5.1","name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":128000},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"qwen3.7-plus":{"id":"qwen3.7-plus","name":"Qwen3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"deepseek-v4-pro":{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"wan2.7-image":{"id":"wan2.7-image","name":"Wan2.7 Image","description":"Image model for prompt-driven generation, editing, and visual design workflows","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-05-29","last_updated":"2026-05-29","modalities":{"input":["text"],"output":["image"]},"open_weights":false,"limit":{"context":8192,"output":0},"cost":{"input":0,"output":0}}}},"neuralwatt":{"id":"neuralwatt","env":["NEURALWATT_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.neuralwatt.com/v1","name":"Neuralwatt","doc":"https://portal.neuralwatt.com/docs","models":{"glm-5.2-short-fast-flex":{"id":"glm-5.2-short-fast-flex","name":"GLM 5.2 Short Fast Flex","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-06-17","last_updated":"2026-06-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":199984,"output":32000},"cost":{"input":0.9425,"output":2.925,"cache_read":0.09425}},"glm-5.2-short-flex":{"id":"glm-5.2-short-flex","name":"GLM 5.2 Short Flex","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-06-17","last_updated":"2026-06-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":199984,"output":32000},"cost":{"input":0.9425,"output":2.925,"cache_read":0.09425}},"glm-5.2-flex":{"id":"glm-5.2-flex","name":"GLM 5.2 Flex","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-06-17","last_updated":"2026-06-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":1048560},"cost":{"input":0.9425,"output":2.925,"cache_read":0.09425}},"glm-5.2":{"id":"glm-5.2","name":"GLM 5.2","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-06-17","last_updated":"2026-06-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":1048560},"cost":{"input":1.45,"output":4.5,"cache_read":0.145}},"deepseek-v4-flash":{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":65536},"cost":{"input":0.14,"output":0.28,"cache_read":0.028}},"kimi-k2.7-code":{"id":"kimi-k2.7-code","name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262128,"output":262128},"cost":{"input":0.95,"output":4,"cache_read":0.095}},"kimi-k2.7-code-flex":{"id":"kimi-k2.7-code-flex","name":"Kimi K2.7 Code Flex","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262128,"output":262128},"cost":{"input":0.6175,"output":2.6,"cache_read":0.06175}},"kimi-k2.7-code-fast":{"id":"kimi-k2.7-code-fast","name":"Kimi K2.7 Code Fast","description":"Kimi K2.7 Code with reasoning capped to a short budget for lower latency; reasoning cannot be disabled on this model","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262128,"output":262128},"cost":{"input":0.95,"output":4,"cache_read":0.095}},"glm-5.2-short-fast":{"id":"glm-5.2-short-fast","name":"GLM 5.2 Short Fast","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-06-17","last_updated":"2026-06-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":199984,"output":32000},"cost":{"input":1.45,"output":4.5,"cache_read":0.145}},"kimi-k3":{"id":"kimi-k3","name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":1048560},"cost":{"input":3,"output":15,"cache_read":0.3}},"kimi-k3-fast":{"id":"kimi-k3-fast","name":"Kimi K3 Fast","description":"Kimi K3 with thinking disabled for low-latency tool calling, vision, and JSON work","family":"kimi-k3","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":1048560},"cost":{"input":3,"output":15,"cache_read":0.3}},"glm-5.2-fast":{"id":"glm-5.2-fast","name":"GLM 5.2 Fast","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-06-17","last_updated":"2026-06-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":1048560},"cost":{"input":1.45,"output":4.5,"cache_read":0.145}},"kimi-k3-flex":{"id":"kimi-k3-flex","name":"Kimi K3 Flex","description":"Kimi K3 on the flex tier: discounted, best-effort latency, requests may be held under load","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":1048560},"cost":{"input":1.95,"output":9.75,"cache_read":0.195}},"qwen3.6-35b-fast":{"id":"qwen3.6-35b-fast","name":"Qwen3.6 35B Fast","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"qwen3.6","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-01","last_updated":"2026-04-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131056,"output":131056},"cost":{"input":0.29,"output":1.15,"cache_read":0.029}},"gemma-4-31b":{"id":"gemma-4-31b","name":"Gemma 4 31B","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","max"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262128,"output":16384},"cost":{"input":0.144,"output":0.42,"cache_read":0.0144}},"deepseek-v4-pro":{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":393216},"status":"beta","cost":{"input":1,"output":3,"cache_read":0.1}},"deepseek-v4-flash-flex":{"id":"deepseek-v4-flash-flex","name":"DeepSeek V4 Flash Flex","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":65536},"cost":{"input":0.091,"output":0.182,"cache_read":0.0182}},"glm-5.3":{"id":"glm-5.3","name":"GLM 5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048560,"output":1048560},"status":"beta","cost":{"input":1.45,"output":4.5,"cache_read":0.145}},"qwen3.6-35b":{"id":"qwen3.6-35b","name":"Qwen3.6 35B","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131056,"output":131056},"cost":{"input":0.29,"output":1.15,"cache_read":0.029}},"qwen-3.8-27b":{"id":"qwen-3.8-27b","name":"Qwen3.8 27B","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262128,"output":65536},"status":"beta","cost":{"input":0.45,"output":3.2,"cache_read":0.25}},"glm-5.2-short":{"id":"glm-5.2-short","name":"GLM 5.2 Short","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"budget_tokens"}],"tool_call":true,"interleaved":true,"structured_output":false,"temperature":true,"release_date":"2026-06-17","last_updated":"2026-06-17","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":199984,"output":32000},"cost":{"input":1.45,"output":4.5,"cache_read":0.145}}}},"abliteration-ai":{"id":"abliteration-ai","env":["ABLIT_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.abliteration.ai/v1","name":"abliteration.ai","doc":"https://docs.abliteration.ai/models","models":{"abliterated-model-large":{"id":"abliterated-model-large","name":"Abliterated Model Large","description":"GLM-5.2 model abliterated and finetuned for cyber, ML red teaming, and agent testing","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-25","last_updated":"2026-07-28","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":999990},"cost":{"input":5,"output":5,"cache_read":0.5}},"abliterated-model-large-v2":{"id":"abliterated-model-large-v2","name":"Abliterated Model Large V2","description":"GLM-5.3 model abliterated and finetuned for cyber, ML red teaming, and agent testing","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-29","last_updated":"2026-08-31","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":1000000,"output":999990},"cost":{"input":5,"output":5,"cache_read":0.5}},"abliterated-model":{"id":"abliterated-model","name":"Abliterated Model","description":"Multimodal model for analyzing text, images, documents, and rich media","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"release_date":"2026-01-06","last_updated":"2026-07-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":150000,"input":150000,"output":8192},"cost":{"input":3,"output":3,"cache_read":0.3}}}},"clarifai":{"id":"clarifai","env":["CLARIFAI_PAT"],"npm":"@ai-sdk/openai-compatible","api":"https://api.clarifai.com/v2/ext/openai/v1","name":"Clarifai","doc":"https://docs.clarifai.com/compute/inference/","models":{"qwen/qwenCoder/models/Qwen3-Coder-30B-A3B-Instruct":{"id":"qwen/qwenCoder/models/Qwen3-Coder-30B-A3B-Instruct","name":"Qwen3 Coder 30B A3B Instruct","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07-31","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.11458,"output":0.74812}},"qwen/qwenLM/models/Qwen3-30B-A3B-Instruct-2507":{"id":"qwen/qwenLM/models/Qwen3-30B-A3B-Instruct-2507","name":"Qwen3 30B A3B Instruct 2507","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-30","last_updated":"2026-02-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.3,"output":0.5}},"qwen/qwenLM/models/Qwen3-30B-A3B-Thinking-2507":{"id":"qwen/qwenLM/models/Qwen3-30B-A3B-Thinking-2507","name":"Qwen3 30B A3B Thinking 2507","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-07-31","last_updated":"2026-02-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":131072},"cost":{"input":0.36,"output":1.3}},"clarifai/main/models/mm-poly-8b":{"id":"clarifai/main/models/mm-poly-8b","name":"MM Poly 8B","description":"Multimodal model for analyzing text, images, documents, and rich media","family":"mm-poly","attachment":true,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2025-06","last_updated":"2026-02-25","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":32768,"output":4096},"cost":{"input":0.658,"output":1.11}},"deepseek-ai/deepseek-ocr/models/DeepSeek-OCR":{"id":"deepseek-ai/deepseek-ocr/models/DeepSeek-OCR","name":"DeepSeek OCR","description":"OCR model for extracting structured text from documents and screenshots","family":"deepseek","attachment":true,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2025-10-20","last_updated":"2026-02-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":8192,"output":8192},"cost":{"input":0.2,"output":0.7}},"mistralai/completion/models/Ministral-3-14B-Reasoning-2512":{"id":"mistralai/completion/models/Ministral-3-14B-Reasoning-2512","name":"Ministral 3 14B Reasoning 2512","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2025-12","release_date":"2025-12-01","last_updated":"2025-12-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":2.5,"output":1.7}},"mistralai/completion/models/Ministral-3-3B-Reasoning-2512":{"id":"mistralai/completion/models/Ministral-3-3B-Reasoning-2512","name":"Ministral 3 3B Reasoning 2512","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2025-12","last_updated":"2026-02-25","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":1.039,"output":0.54825}},"minimaxai/chat-completion/models/MiniMax-M2_5-high-throughput":{"id":"minimaxai/chat-completion/models/MiniMax-M2_5-high-throughput","name":"MiniMax-M2.5 High Throughput","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2}},"openai/chat-completion/models/gpt-oss-120b-high-throughput":{"id":"openai/chat-completion/models/gpt-oss-120b-high-throughput","name":"GPT OSS 120B High Throughput","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2026-02-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":16384},"cost":{"input":0.09,"output":0.36}},"openai/chat-completion/models/gpt-oss-20b":{"id":"openai/chat-completion/models/gpt-oss-20b","name":"GPT OSS 20B","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-12-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":16384},"cost":{"input":0.045,"output":0.18}},"moonshotai/chat-completion/models/Kimi-K2_6":{"id":"moonshotai/chat-completion/models/Kimi-K2_6","name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4}},"arcee_ai/AFM/models/trinity-mini":{"id":"arcee_ai/AFM/models/trinity-mini","name":"Trinity Mini","description":"Reasoning-tuned 26B MoE model with 3B active parameters for agents, tools, and multi-step workloads","family":"trinity","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"knowledge":"2024-10","release_date":"2025-12-01","last_updated":"2026-02-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.045,"output":0.15}}}},"morph":{"id":"morph","env":["MORPH_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.morphllm.com/v1","name":"Morph","doc":"https://docs.morphllm.com/api-reference/introduction","models":{"morph-v3-large":{"id":"morph-v3-large","name":"Morph v3 Large","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"morph","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-08-15","last_updated":"2024-08-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32000,"output":32000},"cost":{"input":0.9,"output":1.9}},"morph-v3-fast":{"id":"morph-v3-fast","name":"Morph v3 Fast","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"morph","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-08-15","last_updated":"2024-08-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":16000,"output":16000},"cost":{"input":0.8,"output":1.2}},"auto":{"id":"auto","name":"Auto","description":"Automatic model router for matching prompts to suitable backends and budgets","family":"auto","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2024-06-01","last_updated":"2024-06-01","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":32000,"output":32000},"cost":{"input":0.85,"output":1.55}}}},"aihubmix":{"id":"aihubmix","env":["AIHUBMIX_API_KEY"],"npm":"@aihubmix/ai-sdk-provider","name":"AIHubMix","doc":"https://docs.aihubmix.com","models":{"claude-sonnet-4-6":{"id":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}}},"qwen3.7-max":{"id":"qwen3.7-max","name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"output":64000},"cost":{"input":1.69,"output":5.07,"cache_read":0.169,"cache_write":2.1125}},"kimi-k2.7-code-highspeed":{"id":"kimi-k2.7-code-highspeed","name":"Kimi K2.7 Code Highspeed","description":"Lower-latency Kimi Code variant for interactive edits and coding-agent loops","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":1.9,"output":7.999,"cache_read":0.32167}},"gemini-3.1-pro-preview-customtools":{"id":"gemini-3.1-pro-preview-customtools","name":"Gemini 3.1 Pro Preview Custom Tools","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"deepseek-v4-pro-0813":{"id":"deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.6918,"output":2.0754,"cache_read":0.023058}},"doubao-seed-2-0-lite-260428":{"id":"doubao-seed-2-0-lite-260428","name":"Doubao Seed 2.0 Lite 260428","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-28","last_updated":"2026-04-28","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":128000},"cost":{"input":0.08,"output":0.51,"cache_read":0.01692,"input_audio":1.269,"tiers":[{"input":0.13,"output":0.76,"cache_read":0.02536,"input_audio":1.902,"tier":{"type":"context","size":32000}},{"input":0.25,"output":1.52,"cache_read":0.05072,"input_audio":3.804,"tier":{"type":"context","size":128000}}]}},"deepseek-v4-flash-0731":{"id":"deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.142,"output":0.284,"cache_read":0.0284}},"coding-minimax-m2.7":{"id":"coding-minimax-m2.7","name":"Coding MiniMax M2.7","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":128100},"cost":{"input":0.2,"output":0.2}},"coding-glm-5.1":{"id":"coding-glm-5.1","name":"Coding GLM 5.1","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-11","last_updated":"2026-04-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":0.06,"output":0.22,"cache_read":0.013}},"claude-opus-4-7-think":{"id":"claude-opus-4-7-think","name":"Claude Opus 4.7 Thinking","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}}},"gpt-5.1-codex-mini":{"id":"gpt-5.1-codex-mini","name":"GPT-5.1 Codex mini","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025}},"grok-4.3":{"id":"grok-4.3","name":"Grok 4.3","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-05-01","last_updated":"2026-05-01","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":1000000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"qwen3.6-plus":{"id":"qwen3.6-plus","name":"Qwen3.6 Plus","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-05-09","last_updated":"2026-05-09","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"output":64000},"cost":{"input":0.28,"output":1.69,"cache_read":0.0282,"cache_write":0.3525,"tiers":[{"input":1.13,"output":6.77,"cache_read":0.1128,"cache_write":1.41,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":1.13,"output":6.77,"cache_read":0.1128,"cache_write":1.41}}},"gpt-5.1-codex":{"id":"gpt-5.1-codex","name":"GPT-5.1 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125}},"doubao-seed-2-0-mini-260428":{"id":"doubao-seed-2-0-mini-260428","name":"Doubao Seed 2.0 Mini 260428","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-28","last_updated":"2026-04-28","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":128000},"cost":{"input":0.03,"output":0.28,"cache_read":0.00564,"input_audio":0.423,"tiers":[{"input":0.06,"output":0.56,"cache_read":0.01128,"input_audio":0.846,"tier":{"type":"context","size":32000}},{"input":0.11,"output":1.13,"cache_read":0.02256,"input_audio":1.692,"tier":{"type":"context","size":128000}}]}},"gpt-5.6-sol":{"id":"gpt-5.6-sol","name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"cache_write":6.25}},"claude-opus-5":{"id":"claude-opus-5","name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"minimax-m2.7":{"id":"minimax-m2.7","name":"MiniMax M2.7","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":128000},"cost":{"input":0.3,"output":1.2,"cache_read":0.06,"cache_write":0.375}},"doubao-seed-2-0-code-preview":{"id":"doubao-seed-2-0-code-preview","name":"Doubao Seed 2.0 Code Preview","description":"Coding model for repository understanding, refactors, and agentic engineering tasks","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":128000},"cost":{"input":0.48,"output":2.41,"cache_read":0.09644,"tiers":[{"input":0.72,"output":3.62,"cache_read":0.144656,"tier":{"type":"context","size":32000}},{"input":1.45,"output":7.23,"cache_read":0.28932,"tier":{"type":"context","size":128000}}]}},"xiaomi-mimo-v2.5-free":{"id":"xiaomi-mimo-v2.5-free","name":"Xiaomi MiMo-V2.5 (free)","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo-v2.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-05-13","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0}},"kimi-k2.6":{"id":"kimi-k2.6","name":"Kimi K2.6","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.95,"output":4,"cache_read":0.16}},"gemini-3.1-pro-preview":{"id":"gemini-3.1-pro-preview","name":"Gemini 3.1 Pro Preview","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"coding-xiaomi-mimo-v2.5-pro":{"id":"coding-xiaomi-mimo-v2.5-pro","name":"Coding Xiaomi MiMo-V2.5-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo-v2.5-pro","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-05-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.2,"output":0.6,"cache_read":0.04,"tiers":[{"input":0.4,"output":1.2,"cache_read":0.08,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.4,"output":1.2,"cache_read":0.08}}},"gpt-5.2-codex":{"id":"gpt-5.2-codex","name":"GPT-5.2 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":128000},"cost":{"input":1.1268,"output":3.9438,"cache_read":0.2817}},"kimi-k2.7-code":{"id":"kimi-k2.7-code","name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.95,"output":3.9995,"cache_read":0.160835}},"grok-4.5":{"id":"grok-4.5","name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-08","last_updated":"2026-07-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":1000000},"cost":{"input":2,"output":6,"cache_read":0.5}},"alicloud-deepseek-v4-pro":{"id":"alicloud-deepseek-v4-pro","name":"DeepSeek V4 Pro (Alibaba Cloud)","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":1.69,"output":3.38,"cache_read":0.13}},"gpt-5.4":{"id":"gpt-5.4","name":"GPT-5.4","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"experimental":{"modes":{"fast":{"cost":{"input":5,"output":30,"cache_read":0.5},"provider":{"body":{"service_tier":"priority"}}}}},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}}},"gemini-3.1-flash-lite":{"id":"gemini-3.1-flash-lite","name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"cache_write":1}},"grok-build-0.1":{"id":"grok-build-0.1","name":"Grok Build 0.1","description":"Fast Grok coding model tuned for agentic engineering and iterative edits","family":"grok-build","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":1,"output":2,"cache_read":0.2}},"xiaomi-mimo-v2.5-pro-free":{"id":"xiaomi-mimo-v2.5-pro-free","name":"Xiaomi MiMo-V2.5-Pro (free)","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo-v2.5-pro","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-05-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0}},"gpt-5.1":{"id":"gpt-5.1","name":"GPT-5.1","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.13}},"deep-deepseek-v4-pro":{"id":"deep-deepseek-v4-pro","name":"DeepSeek V4 Pro (DeepSeek)","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.478,"output":0.956,"cache_read":0.004302}},"deep-deepseek-v4-flash":{"id":"deep-deepseek-v4-flash","name":"DeepSeek V4 Flash (DeepSeek)","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.154,"output":0.308,"cache_read":0.0308}},"claude-opus-4-6":{"id":"claude-opus-4-6","name":"Claude Opus 4.6","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}}},"gemini-3.5-flash":{"id":"gemini-3.5-flash","name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":1.5,"output":9,"cache_read":1.5}},"xiaomi-mimo-v2.5":{"id":"xiaomi-mimo-v2.5","name":"Xiaomi MiMo-V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo-v2.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-05-13","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.44,"output":2.2,"cache_read":0.088,"tiers":[{"input":0.88,"output":4.4,"cache_read":0.176,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.88,"output":4.4,"cache_read":0.176}}},"gpt-5.6-luna":{"id":"gpt-5.6-luna","name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":6,"cache_read":0.1,"cache_write":1.25}},"qwen3.7-flash":{"id":"qwen3.7-flash","name":"Qwen3.7 Flash","description":"Lightweight multimodal Qwen model for high-throughput text, image, and video tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"input":991000,"output":64000},"cost":{"input":0.0282,"output":0.1128,"cache_read":0.00564,"cache_write":0.03525}},"claude-opus-4-7":{"id":"claude-opus-4-7","name":"Claude Opus 4.7","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}}},"alicloud-deepseek-v4-flash":{"id":"alicloud-deepseek-v4-flash","name":"DeepSeek V4 Flash (Alibaba Cloud)","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.028}},"kimi-k3":{"id":"kimi-k3","name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"release_date":"2026-07-16","last_updated":"2026-07-27","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3}},"gpt-5.3-codex":{"id":"gpt-5.3-codex","name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"zai-glm-5.1":{"id":"zai-glm-5.1","name":"GLM-5.1 (Z.ai)","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-03-27","last_updated":"2026-03-27","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":0.845,"output":3.38,"cache_read":0.183112}},"glm-5.3-flash":{"id":"glm-5.3-flash","name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":0.11268,"output":0.39438,"cache_read":0.02817}},"claude-fable-5":{"id":"claude-fable-5","name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":11,"output":55,"cache_read":1.1,"cache_write":13.75}},"qwen3.8-2.4t-a95b":{"id":"qwen3.8-2.4t-a95b","name":"Qwen3.8 2.4T A95B","description":"Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262000,"output":262000},"cost":{"input":2,"output":6,"cache_read":0.5}},"claude-opus-4-8-think":{"id":"claude-opus-4-8-think","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"coding-xiaomi-mimo-v2.5":{"id":"coding-xiaomi-mimo-v2.5","name":"Coding Xiaomi MiMo-V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo-v2.5","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-05-13","modalities":{"input":["text","image","audio","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.08,"output":0.4,"cache_read":0.016,"tiers":[{"input":0.16,"output":0.8,"cache_read":0.032,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.16,"output":0.8,"cache_read":0.032}}},"qwen3.6-flash":{"id":"qwen3.6-flash","name":"Qwen3.6 Flash","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"output":64000},"cost":{"input":0.17,"output":1.01,"cache_read":0.0169,"cache_write":0.21125,"tiers":[{"input":0.68,"output":4.06,"cache_read":0.0676,"cache_write":0.845,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.68,"output":4.06,"cache_read":0.0676,"cache_write":0.845}}},"gpt-5.4-mini":{"id":"gpt-5.4-mini","name":"GPT-5.4 mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"experimental":{"modes":{"fast":{"cost":{"input":1.5,"output":9,"cache_read":0.15},"provider":{"body":{"service_tier":"priority"}}}}},"cost":{"input":0.75,"output":4.5,"cache_read":0.075}},"coding-minimax-m2.7-free":{"id":"coding-minimax-m2.7-free","name":"Coding MiniMax M2.7 (Free)","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax-free","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":128100},"cost":{"input":0,"output":0}},"doubao-seed-2-0-pro":{"id":"doubao-seed-2-0-pro","name":"Doubao Seed 2.0 Pro","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":128000},"cost":{"input":0.48,"output":2.41,"cache_read":0.09644,"tiers":[{"input":0.72,"output":3.62,"cache_read":0.144656,"tier":{"type":"context","size":32000}},{"input":1.45,"output":7.23,"cache_read":0.28932,"tier":{"type":"context","size":128000}}]}},"qwen3.6-max-preview":{"id":"qwen3.6-max-preview","name":"Qwen3.6 Max Preview","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"qwen3.6","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-05-09","last_updated":"2026-05-09","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":240000,"output":64000},"cost":{"input":1.27,"output":7.61,"cache_read":0.1268,"cache_write":1.585,"tiers":[{"input":2.11,"output":12.67,"cache_read":0.2112,"cache_write":2.64,"tier":{"type":"context","size":128000}}]}},"grok-4.6":{"id":"grok-4.6","name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5}},"alicloud-glm-5.1":{"id":"alicloud-glm-5.1","name":"GLM-5.1 (Alibaba Cloud)","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-03-27","last_updated":"2026-03-27","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":0.84,"output":3.38,"cache_read":0.169,"cache_write":1.05625}},"qwen3.8-max":{"id":"qwen3.8-max","name":"Qwen3.8 Max","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"output":128000},"cost":{"input":1.69,"output":5.07,"cache_read":0.169,"cache_write":2.1125}},"kimi-k2.5":{"id":"kimi-k2.5","name":"Kimi K2.5","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":false,"knowledge":"2025-01","release_date":"2026-01","last_updated":"2026-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.6,"output":3,"cache_read":0.1}},"gemini-3-flash-preview":{"id":"gemini-3-flash-preview","name":"Gemini 3 Flash Preview","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"tiers":[{"input":0.5,"output":3,"cache_read":0.05,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":0.5,"output":3,"cache_read":0.05}}},"claude-sonnet-4-6-think":{"id":"claude-sonnet-4-6-think","name":"Claude Sonnet 4.6 Thinking","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}}},"qwen3.7-plus":{"id":"qwen3.7-plus","name":"Qwen3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":991000,"output":64000},"cost":{"input":0.282,"output":1.128,"cache_read":0.0564,"cache_write":0.3525}},"claude-opus-4-8":{"id":"claude-opus-4-8","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"gemini-3.7-flash":{"id":"gemini-3.7-flash","name":"Gemini 3.7 Flash","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075}},"hy3-preview":{"id":"hy3-preview","name":"Hy3 Preview","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-20","last_updated":"2026-04-20","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":256000,"output":128000},"cost":{"input":0.17,"output":0.566661,"cache_read":0.051}},"gemini-2.5-pro":{"id":"gemini-2.5-pro","name":"Gemini 2.5 Pro","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-03-20","last_updated":"2025-06-05","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}}},"gpt-5.6-terra":{"id":"gpt-5.6-terra","name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"cache_write":3.125}},"glm-5.3":{"id":"glm-5.3","name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":128000},"cost":{"input":1.1268,"output":3.9438,"cache_read":0.2817}},"glm-5v-turbo":{"id":"glm-5v-turbo","name":"GLM 5 Vision Turbo","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glmv","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-05-09","last_updated":"2026-05-09","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":128000},"cost":{"input":0.7042,"output":3.09848,"cache_read":0.169008}},"gpt-5.2":{"id":"gpt-5.2","name":"GPT-5.2","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175}},"gemini-2.5-flash":{"id":"gemini-2.5-flash","name":"Gemini 2.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":0,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-03-20","last_updated":"2025-06-05","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"input_audio":1}},"xiaomi-mimo-v2.5-pro":{"id":"xiaomi-mimo-v2.5-pro","name":"Xiaomi MiMo-V2.5-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo-v2.5-pro","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-05-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":1.1,"output":3.3,"cache_read":0.22,"tiers":[{"input":2.2,"output":6.6,"cache_read":0.44,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2.2,"output":6.6,"cache_read":0.44}}},"claude-sonnet-5":{"id":"claude-sonnet-5","name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":true,"structured_output":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"claude-opus-4-6-think":{"id":"claude-opus-4-6-think","name":"Claude Opus 4.6 Thinking","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}}},"coding-glm-5.1-free":{"id":"coding-glm-5.1-free","name":"Coding GLM 5.1 (free)","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm-free","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-11","last_updated":"2026-04-11","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":200000,"output":128000},"cost":{"input":0,"output":0}},"coding-minimax-m2.7-highspeed":{"id":"coding-minimax-m2.7-highspeed","name":"Coding MiniMax M2.7 Highspeed","description":"High-speed MiniMax model for low-latency coding and agent workflows","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":128100},"cost":{"input":0.2,"output":0.2}},"gpt-5.5":{"id":"gpt-5.5","name":"GPT-5.5","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1050000,"input":922000,"output":128000},"experimental":{"modes":{"fast":{"cost":{"input":12.5,"output":75,"cache_read":1.25},"provider":{"body":{"service_tier":"priority"}}}}},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}}}}},"chutes":{"id":"chutes","env":["CHUTES_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://llm.chutes.ai/v1","name":"Chutes","doc":"https://llm.chutes.ai/v1/models","models":{"Nemotron-3-Nano-Omni-30B-TEE":{"id":"Nemotron-3-Nano-Omni-30B-TEE","name":"Nemotron 3 Nano Omni 30B TEE","description":"Omni-modal model for text, vision, audio, and multimodal agent tasks","family":"nemotron","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2026-07-23","last_updated":"2026-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":0},"cost":{"input":0.0245,"output":0.0978,"cache_read":0.0024499999999999995}},"deepseek-ai/DeepSeek-V4-Flash-0731-TEE":{"id":"deepseek-ai/DeepSeek-V4-Flash-0731-TEE","name":"DeepSeek V4 Flash 0731 TEE","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-08-02","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":131072},"cost":{"input":0.44,"output":1.32,"cache_read":0.04399999999999999}},"deepseek-ai/DeepSeek-V3.2-TEE":{"id":"deepseek-ai/DeepSeek-V3.2-TEE","name":"DeepSeek V3.2 TEE","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2024-07","release_date":"2025-12","last_updated":"2026-06-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":65536},"cost":{"input":1,"output":1,"cache_read":0.09999999999999998}},"google/gemma-4-31B-turbo-TEE":{"id":"google/gemma-4-31B-turbo-TEE","name":"gemma 4 31B turbo TEE","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":65536},"cost":{"input":0.12,"output":0.37,"cache_read":0.011999999999999997}},"zai-org/GLM-5.1-TEE":{"id":"zai-org/GLM-5.1-TEE","name":"GLM 5.1 TEE","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-04-07","last_updated":"2026-04-07","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":202752,"output":65535},"cost":{"input":0.98,"output":3.08,"cache_read":0.09799999999999998}},"zai-org/GLM-5.2-TEE":{"id":"zai-org/GLM-5.2-TEE","name":"GLM 5.2 TEE","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":65535},"cost":{"input":1.25,"output":3.95,"cache_read":0.12499999999999997}},"Qwen/Qwen3.8-27B-TEE":{"id":"Qwen/Qwen3.8-27B-TEE","name":"Qwen3.8 27B TEE","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-16","last_updated":"2026-08-16","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.32,"output":2.5,"cache_read":0.031999999999999994}},"Qwen/Qwen3.6-27B-TEE":{"id":"Qwen/Qwen3.6-27B-TEE","name":"Qwen3.6 27B TEE","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.3,"output":2,"cache_read":0.029999999999999992}},"Qwen/Qwen3.5-397B-A17B-TEE":{"id":"Qwen/Qwen3.5-397B-A17B-TEE","name":"Qwen3.5 397B A17B TEE","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.45,"output":3,"cache_read":0.04499999999999999}},"Qwen/Qwen3-235B-A22B-Thinking-2507-TEE":{"id":"Qwen/Qwen3-235B-A22B-Thinking-2507-TEE","name":"Qwen3 235B A22B Thinking 2507 TEE","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-07","last_updated":"2026-06-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":262144},"cost":{"input":0.2989,"output":1.1957,"cache_read":0.029889999999999993}},"Qwen/Qwen3-32B-TEE":{"id":"Qwen/Qwen3-32B-TEE","name":"Qwen3 32B TEE","description":"Dense open Qwen model for self-hosted chat, reasoning, and coding","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-04","last_updated":"2025-04","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":40960,"output":40960},"cost":{"input":0.104,"output":0.416,"cache_read":0.010399999999999998}},"unsloth/Mistral-Nemo-Instruct-2407-TEE":{"id":"unsloth/Mistral-Nemo-Instruct-2407-TEE","name":"Mistral Nemo Instruct 2407 TEE","description":"Efficient Mistral-NVIDIA open model for multilingual chat and local deployment","family":"mistral-nemo","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"knowledge":"2024-07","release_date":"2024-07-01","last_updated":"2024-07-01","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.0245,"output":0.0978,"cache_read":0.0024499999999999995}},"moonshotai/Kimi-K3-TEE":{"id":"moonshotai/Kimi-K3-TEE","name":"Kimi K3 TEE","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-29","last_updated":"2026-07-29","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":65535},"cost":{"input":3,"output":15,"cache_read":0.29999999999999993}},"moonshotai/Kimi-K2.6-TEE":{"id":"moonshotai/Kimi-K2.6-TEE","name":"Kimi K2.6 TEE","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-12","release_date":"2026-04-21","last_updated":"2026-04-21","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65535},"cost":{"input":0.58,"output":3.4,"cache_read":0.05799999999999998}}}},"groq":{"id":"groq","env":["GROQ_API_KEY"],"npm":"@ai-sdk/groq","name":"Groq","doc":"https://console.groq.com/docs/models","models":{"whisper-large-v3":{"id":"whisper-large-v3","name":"Whisper","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"whisper","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2023-09-01","last_updated":"2025-09-05","modalities":{"input":["audio"],"output":["text"]},"open_weights":true,"limit":{"context":0,"output":0}},"whisper-large-v3-turbo":{"id":"whisper-large-v3-turbo","name":"Whisper Large V3 Turbo","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"whisper","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2024-10-01","last_updated":"2024-10-01","modalities":{"input":["audio"],"output":["text"]},"open_weights":true,"limit":{"context":0,"output":0}},"llama-3.1-8b-instant":{"id":"llama-3.1-8b-instant","name":"Llama 3.1 8B","description":"Compact Llama instruction model for fast chat and local deployment","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":131072},"cost":{"input":0.05,"output":0.08}},"allam-2-7b":{"id":"allam-2-7b","name":"ALLaM-2-7b","description":"ALLaM-2-7b instruction tuned model by SDAIA","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2025-01-23","last_updated":"2025-01-23","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":4096,"output":4096},"cost":{"input":0,"output":0}},"llama-3.3-70b-versatile":{"id":"llama-3.3-70b-versatile","name":"Llama 3.3 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.59,"output":0.79}},"qwen/qwen3.8-27b":{"id":"qwen/qwen3.8-27b","name":"Qwen3.8 27B","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","default","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131042,"output":16384},"cost":{"input":0.8,"output":4}},"qwen/qwen3.6-27b":{"id":"qwen/qwen3.6-27b","name":"Qwen3.6 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","default"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-22","last_updated":"2026-04-22","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":16384},"cost":{"input":0.6,"output":3,"cache_read":0.3}},"groq/compound":{"id":"groq/compound","name":"Compound","description":"General-purpose chat model for instruction following, writing, and analysis","family":"groq","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2025-09-04","last_updated":"2025-09-04","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":8192}},"groq/compound-mini":{"id":"groq/compound-mini","name":"Compound Mini","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"groq","attachment":false,"reasoning":false,"tool_call":false,"temperature":true,"release_date":"2025-09-04","last_updated":"2025-09-04","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":8192}},"meta-llama/llama-prompt-guard-2-86m":{"id":"meta-llama/llama-prompt-guard-2-86m","name":"Prompt Guard 2 86M","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-05-29","last_updated":"2025-05-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":512,"output":512},"status":"beta","cost":{"input":0.04,"output":0.04}},"meta-llama/llama-prompt-guard-2-22m":{"id":"meta-llama/llama-prompt-guard-2-22m","name":"Llama Prompt Guard 2 22M","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"llama","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-05-29","last_updated":"2025-05-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":512,"output":512},"status":"beta","cost":{"input":0.03,"output":0.03}},"openai/gpt-oss-20b":{"id":"openai/gpt-oss-20b","name":"GPT OSS 20B","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-09-25","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":65536},"cost":{"input":0.075,"output":0.3,"cache_read":0.0375}},"openai/gpt-oss-safeguard-20b":{"id":"openai/gpt-oss-safeguard-20b","name":"Safety GPT OSS 20B","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-10-29","last_updated":"2026-06-29","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":65536},"status":"beta","cost":{"input":0.075,"output":0.3}},"openai/gpt-oss-120b":{"id":"openai/gpt-oss-120b","name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-08-05","last_updated":"2025-10-21","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":65536},"cost":{"input":0.15,"output":0.6,"cache_read":0.075}},"canopylabs/orpheus-v1-english":{"id":"canopylabs/orpheus-v1-english","name":"Canopy Labs Orpheus V1 English","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"canopylabs","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-12-19","last_updated":"2025-12-19","modalities":{"input":["text"],"output":["audio"]},"open_weights":false,"limit":{"context":4000,"output":50000},"status":"beta"},"canopylabs/orpheus-arabic-saudi":{"id":"canopylabs/orpheus-arabic-saudi","name":"Canopy Labs Orpheus Arabic Saudi","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"canopylabs","attachment":false,"reasoning":false,"tool_call":false,"temperature":false,"release_date":"2025-12-16","last_updated":"2025-12-16","modalities":{"input":["text"],"output":["audio"]},"open_weights":false,"limit":{"context":4000,"output":50000},"status":"beta"}}},"zai-coding-plan":{"id":"zai-coding-plan","env":["ZHIPU_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.z.ai/api/coding/paas/v4","name":"Z.AI Coding Plan","doc":"https://docs.z.ai/devpack/overview","models":{"glm-5.2-highspeed":{"id":"glm-5.2-highspeed","name":"GLM-5.2 Highspeed","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-4.7":{"id":"glm-4.7","name":"GLM-4.7","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-5.3-highspeed":{"id":"glm-5.3-highspeed","name":"GLM-5.3 Highspeed","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-5.3-flash":{"id":"glm-5.3-flash","name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-26","last_updated":"2026-08-26","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-5-turbo":{"id":"glm-5-turbo","name":"GLM-5-Turbo","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-03-16","last_updated":"2026-03-16","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}},"glm-5.3":{"id":"glm-5.3","name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0}}}},"volcengine":{"id":"volcengine","env":["ARK_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://ark.cn-beijing.volces.com/api/v3","name":"Volcengine Ark","doc":"https://www.volcengine.com/docs/82379/1330310","models":{"doubao-seed-2-0-lite-260428":{"id":"doubao-seed-2-0-lite-260428","name":"Seed 2.0 Lite","description":"Cost-efficient ByteDance Seed 2.0 model for production chat, analysis, and structured generation","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":131072},"cost":{"input":0.08906,"output":0.53436,"cache_read":0.01781,"tiers":[{"input":0.13359,"output":0.80154,"cache_read":0.02672,"tier":{"type":"context","size":32000}},{"input":0.26718,"output":1.60308,"cache_read":0.05344,"tier":{"type":"context","size":128000}}]}},"doubao-seed-character-260628":{"id":"doubao-seed-character-260628","name":"Seed Character","description":"ByteDance Seed model optimized for character-driven dialogue and consistent conversational behavior","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-23","last_updated":"2026-06-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.11875,"output":0.29687,"cache_read":0.02375,"tiers":[{"input":0.17812,"output":0.8906,"cache_read":0.02375,"tier":{"type":"context","size":32000}}]}},"doubao-seed-2-0-mini-260428":{"id":"doubao-seed-2-0-mini-260428","name":"Seed 2.0 Mini","description":"Lightweight ByteDance Seed 2.0 model for low-latency multimodal reasoning and high-volume tasks","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":131072},"cost":{"input":0.02969,"output":0.29687,"cache_read":0.00594,"tiers":[{"input":0.05937,"output":0.59374,"cache_read":0.01187,"tier":{"type":"context","size":32000}},{"input":0.11875,"output":1.18747,"cache_read":0.02375,"tier":{"type":"context","size":128000}}]}},"doubao-seed-2-1-pro-260628":{"id":"doubao-seed-2-1-pro-260628","name":"Seed 2.1 Pro","description":"Flagship ByteDance Seed 2.1 model for complex multimodal reasoning, coding, and agents","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-23","last_updated":"2026-06-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.8906,"output":4.45301,"cache_read":0.17812}},"doubao-seed-2-0-code-preview-260215":{"id":"doubao-seed-2-0-code-preview-260215","name":"Seed 2.0 Code","description":"ByteDance Seed coding model for multimodal software engineering and long-running agents","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":131072},"cost":{"input":0.47499,"output":2.37494,"cache_read":0.095,"tiers":[{"input":0.71248,"output":3.56241,"cache_read":0.1425,"tier":{"type":"context","size":32000}},{"input":1.42496,"output":7.12482,"cache_read":0.28499,"tier":{"type":"context","size":128000}}]}},"doubao-seed-1-8-251228":{"id":"doubao-seed-1-8-251228","name":"Seed 1.8","description":"ByteDance Seed model for multimodal reasoning, long-context analysis, and agent workflows","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2025-12-28","last_updated":"2025-12-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":64000},"cost":{"input":0.11875,"output":1.18747,"cache_read":0.02375,"tiers":[{"input":0.17812,"output":2.37494,"cache_read":0.02375,"tier":{"type":"context","size":32000}},{"input":0.35624,"output":3.56241,"cache_read":0.02375,"tier":{"type":"context","size":128000}}]}},"doubao-seed-1-6-flash-250828":{"id":"doubao-seed-1-6-flash-250828","name":"Seed 1.6 Flash","description":"Low-latency ByteDance Seed model for high-throughput chat, extraction, and lightweight tool use","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2025-08-28","last_updated":"2025-08-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000},"cost":{"input":0.02227,"output":0.22265,"cache_read":0.00445,"tiers":[{"input":0.04453,"output":0.4453,"cache_read":0.00445,"tier":{"type":"context","size":32000}},{"input":0.08906,"output":0.8906,"cache_read":0.00445,"tier":{"type":"context","size":128000}}]}},"doubao-seed-1-6-251015":{"id":"doubao-seed-1-6-251015","name":"Seed 1.6","description":"ByteDance Seed model for long-context reasoning, instruction following, and tool-assisted tasks","family":"seed","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":64000},"cost":{"input":0.11875,"output":1.18747,"cache_read":0.02375,"tiers":[{"input":0.17812,"output":2.37494,"cache_read":0.02375,"tier":{"type":"context","size":32000}},{"input":0.35624,"output":3.56241,"cache_read":0.02375,"tier":{"type":"context","size":128000}}]}},"deepseek-v4-pro-ga-260813":{"id":"deepseek-v4-pro-ga-260813","name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":1.3359,"output":4.00771,"cache_read":0.04453}},"doubao-seed-evolving":{"id":"doubao-seed-evolving","name":"Seed Evolving","description":"Rolling ByteDance Seed model for rapidly updated reasoning, coding, and agent capabilities","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-23","last_updated":"2026-06-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.8906,"output":4.45301,"cache_read":0.17812}},"doubao-seed-2-0-pro-260215":{"id":"doubao-seed-2-0-pro-260215","name":"Seed 2.0 Pro","description":"Flagship ByteDance Seed 2.0 model for complex multimodal reasoning and long-horizon agent workflows","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-02-14","last_updated":"2026-02-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":128000},"cost":{"input":0.47499,"output":2.37494,"cache_read":0.095,"tiers":[{"input":0.71248,"output":3.56241,"cache_read":0.1425,"tier":{"type":"context","size":32000}},{"input":1.42496,"output":7.12482,"cache_read":0.28499,"tier":{"type":"context","size":128000}}]}},"doubao-seed-1-6-vision-250815":{"id":"doubao-seed-1-6-vision-250815","name":"Seed 1.6 Vision","description":"ByteDance Seed multimodal model for image understanding, visual reasoning, and tool-assisted tasks","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2025-08-15","last_updated":"2025-08-15","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":32000},"cost":{"input":0.11875,"output":1.18747,"cache_read":0.02375,"tiers":[{"input":0.17812,"output":2.37494,"cache_read":0.02375,"tier":{"type":"context","size":32000}},{"input":0.35624,"output":3.56241,"cache_read":0.02375,"tier":{"type":"context","size":128000}}]}},"glm-5-2-260617":{"id":"glm-5-2-260617","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":1.18747,"output":4.15615,"cache_read":0.29687}},"doubao-seed-2-1-turbo-260628":{"id":"doubao-seed-2-1-turbo-260628","name":"Seed 2.1 Turbo","description":"Faster ByteDance Seed 2.1 model for multimodal reasoning and latency-sensitive agent workflows","family":"seed","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-23","last_updated":"2026-06-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":256000,"output":256000},"cost":{"input":0.4453,"output":2.22651,"cache_read":0.08906}},"deepseek-v4-flash-ga-260731":{"id":"deepseek-v4-flash-ga-260731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.4453,"output":1.3359,"cache_read":0.01484}}}},"sensenova":{"id":"sensenova","env":["SENSENOVA_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://token.sensenova.cn/v1","name":"SenseNova (China)","doc":"https://platform.sensenova.cn/docs","models":{"sensenova-6.8-flash-lite":{"id":"sensenova-6.8-flash-lite","name":"SenseNova 6.8 Flash Lite","description":"SenseNova lightweight multimodal agent model for real-world complex tasks, data analysis, and complex information presentation","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-11","last_updated":"2026-08-28","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":65536},"cost":{"input":0,"output":0,"cache_read":0}},"glm-5.2":{"id":"glm-5.2","name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-06-13","last_updated":"2026-06-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0}},"deepseek-v4-flash":{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":65536},"cost":{"input":0,"output":0,"cache_read":0}}}},"orcarouter":{"id":"orcarouter","env":["ORCAROUTER_API_KEY"],"npm":"@ai-sdk/openai-compatible","api":"https://api.orcarouter.ai/v1","name":"OrcaRouter","doc":"https://docs.orcarouter.ai","models":{"qwen/qwen3.7-max":{"id":"qwen/qwen3.7-max","name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"release_date":"2026-05-21","last_updated":"2026-05-21","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":1.25,"output":3.75,"cache_read":0.25,"cache_write":1.563}},"qwen/qwen3.6-plus":{"id":"qwen/qwen3.6-plus","name":"Qwen3.6 Plus","description":"Earlier Qwen multimodal workhorse for million-token agent and document tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"cache_write":0.625,"tiers":[{"input":2,"output":6,"cache_read":0.2,"cache_write":2.5,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2,"output":6,"cache_read":0.2,"cache_write":2.5}}},"qwen/qwen3.5-27b":{"id":"qwen/qwen3.5-27b","name":"Qwen3.5 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.086,"output":0.688}},"qwen/qwen3.8-27b":{"id":"qwen/qwen3.8-27b","name":"Qwen3.8 27B","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.33,"output":2.4}},"qwen/qwen3.5-35b-a3b":{"id":"qwen/qwen3.5-35b-a3b","name":"Qwen3.5 35B-A3B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.057,"output":0.459}},"qwen/qwen3.5-flash":{"id":"qwen/qwen3.5-flash","name":"Qwen3.5 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":81920}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.1,"output":0.4}},"qwen/qwen3.5-397b-a17b":{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen3.5 397B-A17B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-15","last_updated":"2026-02-15","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.172,"output":1.032}},"qwen/qwen3.7-flash":{"id":"qwen/qwen3.7-flash","name":"Qwen3.7 Flash","description":"Lightweight multimodal Qwen model for high-throughput text, image, and video tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-15","last_updated":"2026-07-15","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"input":991000,"output":65536},"cost":{"input":0.03,"output":0.13,"cache_read":0.006,"cache_write":0.038}},"qwen/qwen3.8-27b-free":{"id":"qwen/qwen3.8-27b-free","name":"Qwen3.8 27B (free)","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-14","last_updated":"2026-08-14","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":65536,"output":32768},"cost":{"input":0,"output":0}},"qwen/qwen3.6-35b-a3b":{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen3.6 35B-A3B","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.248,"output":1.485}},"qwen/qwen3-max":{"id":"qwen/qwen3-max","name":"Qwen3 Max","description":"Flagship Qwen3 model for coding agents, complex reasoning, and tool use","family":"qwen","attachment":false,"reasoning":false,"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":262144,"output":65536},"cost":{"input":0.359,"output":1.434}},"qwen/qwen3.5-122b-a10b":{"id":"qwen/qwen3.5-122b-a10b","name":"Qwen3.5 122B-A10B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-02-23","last_updated":"2026-02-23","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":65536},"cost":{"input":0.115,"output":0.917}},"qwen/qwen3.6-flash":{"id":"qwen/qwen3.6-flash","name":"Qwen3.6 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.6","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-27","last_updated":"2026-04-27","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.25,"output":1.5}},"qwen/qwen3.8-max":{"id":"qwen/qwen3.8-max","name":"Qwen3.8 Max","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens","min":0,"max":262144}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"temperature":true,"release_date":"2026-08-03","last_updated":"2026-08-03","modalities":{"input":["text","image","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25,"cache_write":2.5}},"qwen/qwen3-vl-235b-a22b-thinking":{"id":"qwen/qwen3-vl-235b-a22b-thinking","name":"Qwen3 VL 235B A22B Thinking","description":"Qwen vision-language thinking model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":81920}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":40960},"cost":{"input":0.4,"output":4}},"qwen/qwen3-vl-235b-a22b-instruct":{"id":"qwen/qwen3-vl-235b-a22b-instruct","name":"Qwen3 VL 235B A22B Instruct","description":"Qwen vision-language instruct model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":131072,"output":32768},"cost":{"input":0.4,"output":1.6}},"qwen/qwen3.7-plus":{"id":"qwen/qwen3.7-plus","name":"Qwen3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":0.35,"output":1.42,"cache_read":0.071}},"qwen/qwen3.5-plus":{"id":"qwen/qwen3.5-plus","name":"Qwen3.5 Plus","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"knowledge":"2025-04","release_date":"2026-02-16","last_updated":"2026-02-16","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":65536},"cost":{"input":0.115,"output":0.688,"reasoning":2.4}},"orcarouter/free":{"id":"orcarouter/free","name":"OrcaRouter Free","description":"Built-in router over the free tier that scores each request's difficulty and sends light work to the smaller free model and harder work to the stronger one. Priced at zero and never falls back to a paid model.","family":"auto","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-27","last_updated":"2026-08-27","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":65536,"output":32768},"cost":{"input":0,"output":0}},"orcarouter/fusion-mini":{"id":"orcarouter/fusion-mini","name":"OrcaRouter Fusion Mini","description":"Leaner two-model Fusion panel that runs Claude Opus 4.8 and GPT-5.5 in parallel on hard requests, then has a Claude Opus 4.8 judge return the strongest single answer verbatim. Easy requests fall through to a cheaper default and bill as one call.","family":"model-router","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-06-15","last_updated":"2026-06-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000}},"orcarouter/fusion":{"id":"orcarouter/fusion","name":"OrcaRouter Fusion","description":"Curated fan-out router that runs Claude Opus 4.8, GPT-5.5 and Gemini 3.1 Pro in parallel on hard requests, then has a Claude Opus 4.8 judge return the strongest single answer verbatim. Easy requests fall through to a cheaper default and bill as one call.","family":"model-router","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"release_date":"2026-06-15","last_updated":"2026-06-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000}},"orcarouter/fusion-flash":{"id":"orcarouter/fusion-flash","name":"OrcaRouter Fusion Flash","description":"Budget Fusion panel that runs Gemini 3.5 Flash, MiniMax M2.7 and GLM 5.1 in parallel on hard requests, then has a Claude Opus 4.8 judge return the strongest single answer verbatim. Cost-sensitive fan-out over a 200K window.","family":"model-router","attachment":false,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-06-15","last_updated":"2026-06-15","modalities":{"input":["text"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":128000}},"orcarouter/auto":{"id":"orcarouter/auto","name":"OrcaRouter Auto","description":"Automatic model router for matching prompts to suitable backends and budgets","family":"auto","attachment":true,"reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2025-01-01","last_updated":"2026-05-14","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":128000,"output":16384},"cost":{"input":0,"output":0}},"minimax/minimax-m2.7-highspeed":{"id":"minimax/minimax-m2.7-highspeed","name":"MiniMax-M2.7-highspeed","description":"Low-latency M2.7 variant for interactive coding plans and agent loops","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_read":0.06,"cache_write":0.375}},"minimax/minimax-m2.7":{"id":"minimax/minimax-m2.7","name":"MiniMax-M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-03-18","last_updated":"2026-03-18","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06,"cache_write":0.375}},"minimax/minimax-m2.5":{"id":"minimax/minimax-m2.5","name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-02-12","last_updated":"2026-02-12","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03,"cache_write":0.375}},"minimax/minimax-m3":{"id":"minimax/minimax-m3","name":"MiniMax-M3","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"release_date":"2026-06-01","last_updated":"2026-06-01","modalities":{"input":["text","image","video"],"output":["text"]},"open_weights":true,"limit":{"context":1048576,"output":512000},"cost":{"input":0.3,"output":1.2,"cache_read":0.06}},"minimax/minimax-m2.5-highspeed":{"id":"minimax/minimax-m2.5-highspeed","name":"MiniMax-M2.5-highspeed","description":"High-speed MiniMax model for low-latency coding and agent workflows","family":"minimax","attachment":false,"reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"release_date":"2026-02-13","last_updated":"2026-02-13","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_read":0.03,"cache_write":0.375}},"anthropic/claude-opus-4.8":{"id":"anthropic/claude-opus-4.8","name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-opus-4.7":{"id":"anthropic/claude-opus-4.7","name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"experimental":{"modes":{"fast":{"cost":{"input":30,"output":150,"cache_read":3,"cache_write":37.5},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-opus-5":{"id":"anthropic/claude-opus-5","name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":10}},"anthropic/claude-sonnet-4.6":{"id":"anthropic/claude-sonnet-4.6","name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"anthropic/claude-haiku-4.5":{"id":"anthropic/claude-haiku-4.5","name":"Claude Haiku 4.5 (latest)","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25}},"anthropic/claude-opus-4.6":{"id":"anthropic/claude-opus-4.6","name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"experimental":{"modes":{"fast":{"cost":{"input":30,"output":150,"cache_read":3,"cache_write":37.5},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-fable-5":{"id":"anthropic/claude-fable-5","name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5}},"anthropic/claude-sonnet-4.5":{"id":"anthropic/claude-sonnet-4.5","name":"Claude Sonnet 4.5 (latest)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75}},"anthropic/claude-opus-4.5":{"id":"anthropic/claude-opus-4.5","name":"Claude Opus 4.5 (latest)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25}},"anthropic/claude-sonnet-5":{"id":"anthropic/claude-sonnet-5","name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5}},"google/gemma-4-26b-a4b-it":{"id":"google/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.06,"output":0.33,"cache_read":0.0075}},"google/gemini-3.1-pro-preview-customtools":{"id":"google/gemini-3.1-pro-preview-customtools","name":"Gemini 3.1 Pro Preview Custom Tools","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"google/gemini-3.1-pro-preview":{"id":"google/gemini-3.1-pro-preview","name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"input_audio":2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}}},"google/gemini-2.5-flash-lite":{"id":"google/gemini-2.5-flash-lite","name":"Gemini 2.5 Flash-Lite","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01,"input_audio":0.3}},"google/gemini-3.6-flash":{"id":"google/gemini-3.6-flash","name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":7.5,"cache_read":0.15}},"google/gemini-3.1-flash-lite":{"id":"google/gemini-3.1-flash-lite","name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5}},"google/gemini-3.5-flash":{"id":"google/gemini-3.5-flash","name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"cache_write":0.08333,"input_audio":3}},"google/gemini-3.1-flash-lite-preview":{"id":"google/gemini-3.1-flash-lite-preview","name":"Gemini 3.1 Flash Lite Preview","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-03-03","last_updated":"2026-03-03","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5}},"google/gemini-3.5-flash-lite":{"id":"google/gemini-3.5-flash-lite","name":"Gemini 3.5 Flash Lite","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03}},"google/gemini-robotics-er-1.6-preview":{"id":"google/gemini-robotics-er-1.6-preview","name":"Gemini Robotics-ER 1.6 Preview","description":"Vision-language model for embodied reasoning: spatial understanding, task planning, and physical-world agentic robotics","family":"gemini","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2026-04-14","last_updated":"2026-04-14","modalities":{"input":["text","image","video","audio"],"output":["text"]},"open_weights":false,"limit":{"context":131072,"output":65536},"cost":{"input":1,"output":5}},"google/gemini-flash-lite-latest":{"id":"google/gemini-flash-lite-latest","name":"Gemini Flash-Lite Latest","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5}},"google/gemma-4-31b-it":{"id":"google/gemma-4-31b-it","name":"Gemma 4 31B IT","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","attachment":true,"reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-02","last_updated":"2026-04-02","modalities":{"input":["text","image"],"output":["text"]},"open_weights":true,"limit":{"context":262144,"output":32768},"cost":{"input":0.13,"output":0.38,"cache_read":0.02}},"google/gemini-3-flash-preview":{"id":"google/gemini-3-flash-preview","name":"Gemini 3 Flash Preview","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"input_audio":1}},"google/gemini-2.5-pro":{"id":"google/gemini-2.5-pro","name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}}},"google/gemini-flash-latest":{"id":"google/gemini-flash-latest","name":"Gemini Flash Latest","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.1}},"google/gemini-2.5-flash":{"id":"google/gemini-2.5-flash","name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"input_audio":1}},"grok/grok-4.3":{"id":"grok/grok-4.3","name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-17","last_updated":"2026-04-17","modalities":{"input":["text","image","pdf"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}}},"grok/grok-4.5":{"id":"grok/grok-4.5","name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-07-08","last_updated":"2026-07-08","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5}},"grok/grok-4.6":{"id":"grok/grok-4.6","name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5}},"meta/muse-spark-1.2":{"id":"meta/muse-spark-1.2","name":"Muse Spark 1.2","description":"Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-08-05","last_updated":"2026-08-05","modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"meta/muse-spark-1.1":{"id":"meta/muse-spark-1.1","name":"Muse Spark 1.1","description":"Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.","family":"muse","attachment":true,"reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"release_date":"2026-04-08","last_updated":"2026-07-09","modalities":{"input":["text","image","pdf","video"],"output":["text"]},"open_weights":false,"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15}},"deepseek/deepseek-v4-flash-vision-exp":{"id":"deepseek/deepseek-v4-flash-vision-exp","name":"DeepSeek V4 Flash Vision Exp","description":"Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work","family":"deepseek-flash","attachment":true,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-21","last_updated":"2026-08-21","modalities":{"input":["text","image"],"output":["text"]},"open_weights":false,"limit":{"context":1000000,"output":384000},"cost":{"input":0.147,"output":0.295,"cache_read":0.02}},"deepseek/deepseek-v4-pro-0813":{"id":"deepseek/deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"release_date":"2026-08-12","last_updated":"2026-08-22","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.442,"output":0.884,"cache_read":0.06}},"deepseek/deepseek-v4-flash-0731":{"id":"deepseek/deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0.147,"output":0.295,"cache_read":0.02}},"deepseek/deepseek-v4-flash-free":{"id":"deepseek/deepseek-v4-flash-free","name":"DeepSeek V4 Flash (free)","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","attachment":false,"reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"interleaved":{"field":"reasoning_content"},"structured_output":true,"temperature":true,"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","modalities":{"input":["text"],"output":["text"]},"open_weights":true,"limit":{"context":1000000,"output":384000},"cost":{"input":0,"output":0}},"deepseek/deepseek-v4-flaFile truncated at 1 MB.