A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.
Python artificial-intelligencedagdata-sciencedata-visualizationdataflowdeveloper-toolsmachine-learningnotebookspipelinepythonreactivesqlweb-app
Update LLM model catalog (#11002)
github-actions[bot]e32a63b26aa7
diff --git a/packages/llm-info/data/models.yml b/packages/llm-info/data/models.ymlindex 63199e7ddf..476ba14a85 100644--- a/packages/llm-info/data/models.yml+++ b/packages/llm-info/data/models.yml@@ -1,8 +1,18 @@anthropic:+ - name: Claude Opus 5.5+ model: claude-opus-5-5+ description: "Anthropic's strongest Opus model for coding, agents, and knowledge work"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image, pdf]+ output_types: [text]+ release_date: 2026-09-22+ cost: {input: 4, output: 20}+- name: Claude Fable 5.1model: claude-fable-5-1- description: "Anthropic's most capable generally available model for ambitious, long-running coding, agentic, research, and enterprise workflows"+ description: "Anthropic model for ambitious, long-running coding, agentic, research, and enterprise workflows"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]@@ -50,16 +60,6 @@ anthropic:release_date: 2026-05-28cost: {input: 5, output: 25}- - name: Claude Opus 4.7- model: claude-opus-4-7- description: "Anthropic Opus model built for long-running, asynchronous agents"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-04-16- cost: {input: 5, output: 25}-- name: Claude Sonnet 4.6model: claude-sonnet-4-6description: "Anthropic Sonnet model for coding, agents, and professional work"@@ -70,47 +70,57 @@ anthropic:release_date: 2026-02-17cost: {input: 3, output: 15}- - name: Claude Opus 4.6- model: claude-opus-4-6- description: "Anthropic Opus model for coding and long-running professional tasks"+ - name: Claude Haiku 4.5+ model: claude-haiku-4-5+ description: "Anthropic's fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]output_types: [text]- release_date: 2026-02-05- cost: {input: 5, output: 25}+ release_date: 2025-10-15+ cost: {input: 1, output: 5}- - name: Claude Opus 4.5- model: claude-opus-4-5- description: "Anthropic Opus reasoning model for complex software engineering, agentic workflows, and long-horizon computer use"+openai:+ - name: GPT-6 Luna+ model: gpt-6-luna+ description: "OpenAI's most efficient model for focused, high-volume tasks"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]output_types: [text]- release_date: 2025-11-24- cost: {input: 5, output: 25}+ release_date: 2026-09-22+ cost: {input: 0.1, output: 0.5}- - name: Claude Haiku 4.5- model: claude-haiku-4-5- description: "Anthropic's fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models"+ - name: GPT-6 Sol+ model: gpt-6-sol+ description: "OpenAI reasoning model for demanding tasks that balances capability, latency, and cost"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]output_types: [text]- release_date: 2025-10-15- cost: {input: 1, output: 5}+ release_date: 2026-09-22+ cost: {input: 2, output: 10}- - name: Claude Sonnet 4.5- model: claude-sonnet-4-5- description: "Anthropic Sonnet model optimized for real-world agents and coding workflows"+ - name: Daybreak Blue+ model: gpt-daybreak-blue-latest+ description: "OpenAI model alias for authorized defensive cybersecurity work with reduced refusals"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]output_types: [text]- release_date: 2025-09-29- cost: {input: 3, output: 15}+ release_date: 2026-08-07+ cost: {input: 4, output: 20}++ - name: Daybreak Red+ model: gpt-daybreak-red-latest+ description: "OpenAI model alias for advanced authorized cybersecurity work, including penetration testing and exploit validation"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image]+ output_types: [text]+ release_date: 2026-08-07+ cost: {input: 12.5, output: 75}-openai:- name: GPT-6 Astramodel: gpt-6-astradescription: "OpenAI model for advanced analysis, software engineering, deep research, scientific work, and document creation"@@ -171,56 +181,17 @@ openai:release_date: 2026-04-23cost: {input: 5, output: 30}- - name: GPT-5.4 mini- model: gpt-5.4-mini- description: "GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-03-17- cost: {input: 0.75, output: 4.5}-- - name: GPT-5.4 nano- model: gpt-5.4-nano- description: "Most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-03-17- cost: {input: 0.2, output: 1.25}-- - name: GPT-5.4- model: gpt-5.4- description: "OpenAI model that unified the Codex and GPT lines into a single system"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-03-05- cost: {input: 2.5, output: 15}-- - name: GPT-5.3 Codex- model: gpt-5.3-codex- description: "Agentic coding model that combines GPT-5.2-Codex software engineering performance with GPT-5.2 reasoning and professional knowledge"+google:+ - name: Claude Opus 5.5+ model: claude-opus-5-5@default+ description: "Anthropic's strongest Opus model for coding, agents, and knowledge work, available through Vertex AI only"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]output_types: [text]- release_date: 2026-02-05- cost: {input: 1.75, output: 14}-- - name: ChatGPT Image- model: chatgpt-image-latest- description: "Previous-generation image model used in ChatGPT for image generation and editing from text and image inputs"- roles: [chat, edit]- capabilities: []- input_types: [text, image]- output_types: [text, image]- release_date: 2025-12-16+ release_date: 2026-09-22+ cost: {input: 4, output: 20}-google:- name: Grok 4.6model: xai/grok-4.6description: "xAI frontier model for coding, knowledge work, and STEM"@@ -243,7 +214,7 @@ google:- name: Claude Fable 5.1model: claude-fable-5-1@default- description: "Anthropic's most capable generally available model for ambitious, long-running coding, agentic, research, and enterprise workflows, available through Vertex AI only"+ description: "Anthropic model for ambitious, long-running coding, agentic, research, and enterprise workflows, available through Vertex AI only"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]@@ -271,46 +242,6 @@ google:release_date: 2026-08-13cost: {input: 0.75, output: 3.75}- - name: Gemini 3.6 Flash- model: gemini-3.6-flash- description: "Google's high-efficiency model for coding, agentic workflows, and web and app development"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-07-21- cost: {input: 1.5, output: 7.5}-- - name: Gemini 3.5 Flash Lite- model: gemini-3.5-flash-lite- description: "Google's high-efficiency model with upgraded agentic capabilities, suited for focused subagent tasks in multi-agent workflows"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-07-21- cost: {input: 0.3, output: 2.5}-- - name: Gemini 3.5 Flash- model: gemini-3.5-flash- description: "Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-05-19- cost: {input: 1.5, output: 9}-- - name: Gemini 3.1 Flash Lite- model: gemini-3.1-flash-lite- description: "Google's GA high-efficiency multimodal model optimized for low-latency, high-volume workloads"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-05-07- cost: {input: 0.25, output: 1.5}-- name: Gemma 4 26B A4B ITmodel: gemma-4-26b-a4b-itdescription: "Instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind"@@ -349,26 +280,6 @@ google:release_date: 2026-05-28cost: {input: 0.5, output: 60}- - name: Gemini 3.1 Pro Preview- model: gemini-3.1-pro-preview- description: "Google's frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-02-19- cost: {input: 2, output: 12}-- - name: Gemini 3.1 Pro Preview Custom Tools- model: gemini-3.1-pro-preview-customtools- description: "Variant of Gemini 3.1 Pro with refined tool selection that avoids overusing a generic bash tool"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-02-19- cost: {input: 2, output: 12}-- name: Claude Opus 5model: claude-opus-5@defaultdescription: "Anthropic Claude Opus 5 on Google Vertex AI only (not available via the Gemini API)"@@ -400,6 +311,26 @@ google:cost: {input: 5, output: 25}bedrock:+ - name: Claude Opus 5.5 (Global)+ model: global.anthropic.claude-opus-5-5+ description: "Anthropic's strongest Opus model for coding, agents, and knowledge work"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image, pdf]+ output_types: [text]+ release_date: 2026-09-22+ cost: {input: 4, output: 20}++ - name: GPT-6 Luna (Global)+ model: global.openai.gpt-6-luna+ description: "OpenAI's most efficient model for focused, high-volume tasks"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image]+ output_types: [text]+ release_date: 2026-09-22+ cost: {input: 0.1, output: 0.5}+- name: GPT-6 Astra (Global)model: global.openai.gpt-6-astradescription: "OpenAI model for advanced analysis, software engineering, deep research, scientific work, and document creation"@@ -422,7 +353,7 @@ bedrock:- name: Claude Fable 5.1 (Global)model: global.anthropic.claude-fable-5-1- description: "Anthropic's most capable generally available model for ambitious, long-running coding, agentic, research, and enterprise workflows"+ description: "Anthropic model for ambitious, long-running coding, agentic, research, and enterprise workflows"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]@@ -550,36 +481,6 @@ bedrock:release_date: 2026-05-28cost: {input: 5, output: 25}- - name: Claude Opus 4.7- model: anthropic.claude-opus-4-7- description: "Anthropic Opus model built for long-running, asynchronous agents"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-04-16- cost: {input: 5, output: 25}-- - name: Claude Opus 4.7 (Global)- model: global.anthropic.claude-opus-4-7- description: "Anthropic Opus model built for long-running, asynchronous agents"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-04-16- cost: {input: 5, output: 25}-- - name: MiniMax M2.5- model: minimax.minimax-m2.5- description: "Large language model designed for real-world productivity"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text]- output_types: [text]- release_date: 2026-03-18- cost: {input: 0.3, output: 1.2}-- name: GLM-5model: zai.glm-5description: "Z.ai's flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows"@@ -611,6 +512,56 @@ bedrock:cost: {input: 3, output: 15}azure:+ - name: Claude Opus 5.5+ model: claude-opus-5-5+ description: "Anthropic's strongest Opus model for coding, agents, and knowledge work"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image, pdf]+ output_types: [text]+ release_date: 2026-09-22+ cost: {input: 4, output: 20}++ - name: GPT-6 Luna+ model: gpt-6-luna+ description: "OpenAI's most efficient model for focused, high-volume tasks"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image, pdf]+ output_types: [text]+ release_date: 2026-09-22+ cost: {input: 0.1, output: 0.5}++ - name: GPT-6 Sol+ model: gpt-6-sol+ description: "OpenAI reasoning model for demanding tasks that balances capability, latency, and cost"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image, pdf]+ output_types: [text]+ release_date: 2026-09-22+ cost: {input: 2, output: 10}++ - name: GPT Image 2.5 Flare+ model: gpt-image-2.5-flare+ description: "Small image generation and editing model optimized for speed and high-quality everyday use"+ roles: [chat, edit]+ capabilities: []+ input_types: [text, image]+ output_types: [image]+ release_date: 2026-09-08+ cost: {input: 5, output: 30}++ - name: GPT Image 2.5 Sunburst+ model: gpt-image-2.5-sunburst+ description: "Image generation and editing model optimized for quality, precise editing, and subject preservation"+ roles: [chat, edit]+ capabilities: []+ input_types: [text, image]+ output_types: [image]+ release_date: 2026-09-08+ cost: {input: 5, output: 30}+- name: GPT-6 Astramodel: gpt-6-astradescription: "OpenAI model for advanced analysis, software engineering, deep research, scientific work, and document creation"@@ -623,7 +574,7 @@ azure:- name: Claude Fable 5.1model: claude-fable-5-1- description: "Anthropic's most capable generally available model for ambitious, long-running coding, agentic, research, and enterprise workflows"+ description: "Anthropic model for ambitious, long-running coding, agentic, research, and enterprise workflows"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image, pdf]@@ -720,16 +671,6 @@ azure:release_date: 2026-04-24cost: {input: 5, output: 30}- - name: Kimi K2.6- model: kimi-k2.6- description: "Moonshot AI multimodal model for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-04-22- cost: {input: 0.95, output: 4}-- name: Grok 4.20 (Non-Reasoning)model: grok-4-20-non-reasoningdescription: "Reasoning model from xAI with industry-leading speed and agentic tool calling capabilities"@@ -750,46 +691,6 @@ azure:release_date: 2026-04-08cost: {input: 2, output: 6}- - name: GPT-5.4 Mini- model: gpt-5.4-mini- description: "GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-03-17- cost: {input: 0.75, output: 4.5}-- - name: GPT-5.4 Nano- model: gpt-5.4-nano- description: "Most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-03-17- cost: {input: 0.2, output: 1.25}-- - name: GPT-5.4- model: gpt-5.4- description: "OpenAI model that unified the Codex and GPT lines into a single system"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-03-05- cost: {input: 2.5, output: 15}-- - name: GPT-5.3 Codex- model: gpt-5.3-codex- description: "Agentic coding model that combines GPT-5.2-Codex software engineering performance with GPT-5.2 reasoning and professional knowledge"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-02-24- cost: {input: 1.75, output: 14}-- name: Claude Opus 4.8model: claude-opus-4-8description: "Modest but tangible improvement over Claude Opus 4.7"@@ -903,24 +804,6 @@ ollama:output_types: [text]release_date: 2026-04-24- - name: Kimi K2.7 Code- model: kimi-k2.7-code- description: "Coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-06-12-- - name: Kimi K2.6- model: kimi-k2.6- description: "Moonshot AI multimodal model for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-04-20-- name: Gemma 4 31Bmodel: gemma4:31bdescription: "Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output"@@ -939,15 +822,6 @@ ollama:output_types: [text]release_date: 2026-06-13- - name: GLM 5.1- model: glm-5.1- description: "GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text]- output_types: [text]- release_date: 2026-03-27-- name: MiniMax M3model: minimax-m3description: "Multimodal foundation model from MiniMax with a 1M-token context window, suited for long-horizon agentic work and coding"@@ -994,6 +868,16 @@ ollama:release_date: 2026-02-15wandb:+ - name: DeepSeek V4.1 Flash+ model: deepseek-ai/DeepSeek-V4.1-Flash+ description: "Sparse Mixture-of-Experts model from DeepSeek built on a Causal Encoder-Decoder architecture"+ roles: [chat, edit]+ capabilities: [thinking, tool_calling]+ input_types: [text, image]+ output_types: [text]+ release_date: 2026-09-10+ cost: {input: 0.2, output: 0.65}+- name: GLM 5.3 Flashmodel: zai-org/GLM-5.3-Flashdescription: "Native multimodal model from Z.ai optimized for efficient coding and long-horizon agent tasks"@@ -1064,16 +948,6 @@ wandb:release_date: 2026-06-12cost: {input: 0.23, output: 0.96}- - name: Granite 4.1 8B- model: ibm-granite/granite-4.1-8b- description: "Dense 8B decoder-only model from IBM's Granite 4.1 family with a 131K-token context window, designed for enterprise tasks"- roles: [chat, edit]- capabilities: [tool_calling]- input_types: [text]- output_types: [text]- release_date: 2026-04-29- cost: {input: 0.05, output: 0.1}-- name: DeepSeek V4-Flashmodel: deepseek-ai/DeepSeek-V4-Flashdescription: "DeepSeek V4-Flash is an MoE model with 1M context length great for coding, reasoning, and agentic workloads"@@ -1104,16 +978,6 @@ wandb:release_date: 2026-06-16cost: {input: 0.76, output: 2.42}- - name: Z.AI GLM-5.1- model: zai-org/GLM-5.1- description: "GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text]- output_types: [text]- release_date: 2026-04-07- cost: {input: 1.4, output: 4.4}-- name: Nemotron 3 Ultramodel: nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55Bdescription: "Open reasoning and orchestration model from NVIDIA with 55B active parameters out of 550B total (MoE)"@@ -1134,87 +998,87 @@ wandb:release_date: 2026-03-11cost: {input: 0.2, output: 0.8}- - name: MiniMax M2.5- model: MiniMaxAI/MiniMax-M2.5- description: "Large language model designed for real-world productivity"+ - name: gpt-oss-120b+ model: openai/gpt-oss-120b+ description: "Open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text]output_types: [text]- release_date: 2026-02-12- cost: {input: 0.3, output: 1.2}+ release_date: 2025-08-05+ cost: {input: 0.04, output: 0.14}- - name: Kimi K2.7 Code- model: moonshotai/Kimi-K2.7-Code- description: "Kimi K2.7 Code is a 1T-parameter MoE model with 32B active parameters purpose-built for long-horizon agentic coding and software engineering"+ - name: gpt-oss-20b+ model: openai/gpt-oss-20b+ description: "Open-weight 21B parameter model released by OpenAI under the Apache 2.0 license"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text]output_types: [text]- release_date: 2026-06-12- cost: {input: 0.71, output: 3.5}+ release_date: 2025-08-05+ cost: {input: 0.03, output: 0.13}- - name: Moonshot AI Kimi K2.6- model: moonshotai/Kimi-K2.6- description: "Kimi K2.6 is a multimodal Mixture-of-Experts language model featuring 32 billion activated parameters and a total of 1 trillion parameters"+ - name: Mellum2 12B A2.5B+ model: JetBrains/Mellum2-12B-A2.5B-Instruct+ description: "Mellum2-12B-A2.5B-Instruct is a fast MoE model with 131K context built for coding, tool use, and low-latency AI workflows"roles: [chat, edit]capabilities: [thinking, tool_calling]- input_types: [text, image]+ input_types: [text]output_types: [text]- release_date: 2026-04-20- cost: {input: 0.95, output: 4}+ release_date: 2026-06-01+ cost: {input: 0.05, output: 0.1}- - name: gpt-oss-120b- model: openai/gpt-oss-120b- description: "Open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases"+opencode-go:+ - name: LongCat 2.5 Preview Free+ model: longcat-2.5-preview-free+ description: "Multimodal reasoning model from Meituan for coding and agentic workflows, with a one-million-token context window"roles: [chat, edit]capabilities: [thinking, tool_calling]- input_types: [text]+ input_types: [text, image]output_types: [text]- release_date: 2025-08-05- cost: {input: 0.04, output: 0.14}+ release_date: 2026-09-25+ cost: {input: 0, output: 0}- - name: gpt-oss-20b- model: openai/gpt-oss-20b- description: "Open-weight 21B parameter model released by OpenAI under the Apache 2.0 license"+ - name: Space Bunny Free+ model: space-bunny-free+ description: "Anonymous model with fast inference, strong coding capabilities, native multimodal input, and adjustable reasoning effort"roles: [chat, edit]capabilities: [thinking, tool_calling]- input_types: [text]+ input_types: [text, image]output_types: [text]- release_date: 2025-08-05- cost: {input: 0.03, output: 0.13}+ release_date: 2026-09-23+ cost: {input: 0, output: 0}- - name: Qwen3.6 35B A3B- model: Qwen/Qwen3.6-35B-A3B- description: "Qwen3.6-35B-A3B is an MoE multimodal model with 262K context optimized for agentic coding workflows"+ - name: GPT-6 Luna+ model: gpt-6-luna+ description: "OpenAI's most efficient model for focused, high-volume tasks"roles: [chat, edit]capabilities: [thinking, tool_calling]- input_types: [text, image]+ input_types: [text, image, pdf]output_types: [text]- release_date: 2026-04-16- cost: {input: 0.25, output: 1.25}+ release_date: 2026-09-22+ cost: {input: 0.1, output: 0.5}- - name: Qwen3.6 27B- model: Qwen/Qwen3.6-27B- description: "Qwen3.6-27B is a 27B dense multimodal model with 262K context built for flagship-level agentic coding"+ - name: MiMo-V2.6-Flash+ model: mimo-v2.6-flash+ description: "Open-source Mixture-of-Experts foundation model from Xiaomi with 309B total parameters and 15B activated per token"roles: [chat, edit]capabilities: [thinking, tool_calling]input_types: [text, image]output_types: [text]- release_date: 2026-04-22- cost: {input: 0.6, output: 3.6}+ release_date: 2026-09-22+ cost: {input: 0.14, output: 0.28}- - name: Mellum2 12B A2.5B- model: JetBrains/Mellum2-12B-A2.5B-Instruct- description: "Mellum2-12B-A2.5B-Instruct is a fast MoE model with 131K context built for coding, tool use, and low-latency AI workflows"+ - name: MiMo-V2.6-Pro+ model: mimo-v2.6-pro+ description: "Xiaomi's flagship foundation model with over one trillion parameters, designed for demanding coding and agentic tasks"roles: [chat, edit]capabilities: [thinking, tool_calling]- input_types: [text]+ input_types: [text, image]output_types: [text]- release_date: 2026-06-01- cost: {input: 0.05, output: 0.1}+ release_date: 2026-09-22+ cost: {input: 0.435, output: 0.87}-opencode-go:- name: DeepSeek V4.1 Flashmodel: deepseek-v4.1-flashdescription: "Sparse Mixture-of-Experts model from DeepSeek built on a Causal Encoder-Decoder architecture"@@ -1275,16 +1139,6 @@ opencode-go:release_date: 2026-08-21cost: {input: 0.22, output: 0.66}- - name: Muse Spark 1.2 Contributor- model: muse-spark-1.2-contributor- description: "Discounted Meta coding model that permits prompts and completions to train future Meta models"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image, pdf]- output_types: [text]- release_date: 2026-08-05- cost: {input: 0.1, output: 0.2}-- name: GLM-5.3model: glm-5.3description: "Flagship GLM model for long-horizon coding, agents, and complex project delivery"@@ -1345,26 +1199,6 @@ opencode-go:release_date: 2026-04-24cost: {input: 1.74, output: 3.48}- - name: Kimi K2.7 Code- model: kimi-k2.7-code- description: "Coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-06-12- cost: {input: 0.95, output: 4}-- - name: Kimi K2.6- model: kimi-k2.6- description: "Moonshot AI multimodal model for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-04-21- cost: {input: 0.95, output: 4}-- name: GLM 5.2model: glm-5.2description: "Large-scale reasoning model from Z.ai with a 1M-token context window, suited for long-horizon agent workflows and project-level software engineering"@@ -1404,23 +1238,3 @@ opencode-go:output_types: [text]release_date: 2026-05-31cost: {input: 0.1, output: 0.4}-- - name: MiMo V2.5- model: mimo-v2.5- description: "Native omnimodal model from Xiaomi delivering Pro-level agentic performance at roughly half the inference cost"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text, image]- output_types: [text]- release_date: 2026-04-22- cost: {input: 0.4, output: 2}-- - name: MiMo V2.5 Pro- model: mimo-v2.5-pro- description: "Xiaomi's flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as ClawEval, GDPVal,…"- roles: [chat, edit]- capabilities: [thinking, tool_calling]- input_types: [text]- output_types: [text]- release_date: 2026-04-22- cost: {input: 1, output: 3}