Skip to content

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.

Update LLM model catalog (#11183)

github-actions[bot]7c2ee9d40c25
diff --git a/packages/llm-info/data/models.yml b/packages/llm-info/data/models.yml
index b342593841..f3390211a4 100644
--- a/packages/llm-info/data/models.yml
+++ b/packages/llm-info/data/models.yml
@@ -1,5 +1,20 @@
anthropic:
+ - name: Claude Haiku 5.5
+ model: claude-haiku-5-5
+ description: "Small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-10-07
+ cost: {input: 0.1, output: 0.5}
+ reasoning_options:
+ - type: toggle
+ - type: effort
+ values: [low, medium, high, xhigh, max]
+ limits: {context: 1000000, output: 128000}
+
- name: Claude Sonnet 5.5
model: claude-sonnet-5-5
description: "Sonnet-class reasoning model for everyday coding, bug fixes, and document creation"
@@ -117,7 +132,7 @@ anthropic:
- 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"
+ description: "Fast and efficient Claude 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]
@@ -285,6 +300,49 @@ openai:
limits: {context: 1050000, output: 128000}
google:
+ - name: Gemini 4 Argon
+ model: gemini-4-argon
+ description: "Reasoning model for complex software engineering, mathematical and scientific problem solving, and autonomous multi-step agents"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-10-13
+ cost: {input: 2, output: 10}
+ reasoning_options:
+ - type: effort
+ values: [low, medium, high]
+ limits: {context: 1048576, output: 32768}
+
+ - name: Claude Haiku 5.5
+ model: claude-haiku-5-5@default
+ description: "Small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use, available through Vertex AI only"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-10-07
+ cost: {input: 0.1, output: 0.5}
+ reasoning_options:
+ - type: toggle
+ - type: effort
+ values: [low, medium, high, xhigh, max]
+ limits: {context: 1000000, output: 128000}
+
+ - name: Grok 4.7
+ model: xai/grok-4.7
+ description: "Model for coding, agentic tasks, and knowledge work, available through Vertex AI only"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image]
+ output_types: [text]
+ release_date: 2026-09-21
+ cost: {input: 2, output: 6}
+ reasoning_options:
+ - type: effort
+ values: [low, medium, high]
+ limits: {context: 524288, output: 500000}
+
- name: Claude Sonnet 5.5
model: claude-sonnet-5-5@default
description: "Sonnet-class reasoning model for everyday coding, bug fixes, and document creation, available through Vertex AI only"
@@ -322,7 +380,9 @@ google:
output_types: [text]
release_date: 2026-08-12
cost: {input: 2, output: 6}
- reasoning_options: []
+ reasoning_options:
+ - type: effort
+ values: [low, medium, high]
limits: {context: 524288, output: 500000}
- name: Gemini 3.8 Flash
@@ -405,34 +465,6 @@ google:
- type: toggle
limits: {context: 262144, output: 32768}
- - name: Nano Banana 2 Lite
- model: gemini-3.1-flash-lite-image
- description: "Google's fastest, most cost-efficient Gemini image model for high-velocity text-to-image generation and visual exploration"
- roles: [chat, edit]
- capabilities: [thinking, tool_calling]
- input_types: [text, image]
- output_types: [text, image]
- release_date: 2026-06-30
- cost: {input: 0.25, output: 30}
- reasoning_options:
- - type: effort
- values: [minimal, high]
- limits: {context: 65536, output: 4096}
-
- - name: Nano Banana 2
- model: gemini-3.1-flash-image
- description: "Google's image generation and editing model, delivering Pro-level visual quality at Flash speed"
- roles: [chat, edit]
- capabilities: [thinking]
- input_types: [text, image, pdf]
- output_types: [text, image]
- release_date: 2026-05-28
- cost: {input: 0.5, output: 60}
- reasoning_options:
- - type: effort
- values: [minimal, high]
- limits: {context: 131072, output: 32768}
-
- name: Claude Opus 5
model: claude-opus-5@default
description: "Anthropic Claude Opus 5 on Google Vertex AI only (not available via the Gemini API)"
@@ -477,6 +509,21 @@ google:
limits: {context: 1000000, output: 128000}
bedrock:
+ - name: Claude Haiku 5.5 (Global)
+ model: global.anthropic.claude-haiku-5-5
+ description: "Small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-10-07
+ cost: {input: 0.1, output: 0.5}
+ reasoning_options:
+ - type: toggle
+ - type: effort
+ values: [low, medium, high, xhigh, max]
+ limits: {context: 1000000, output: 128000}
+
- name: GPT-6.1 Sol (Global)
model: global.openai.gpt-6.1-sol
description: "Reasoning model for agentic coding, computer use, and document-heavy professional work"
@@ -776,6 +823,21 @@ bedrock:
limits: {context: 1000000, output: 128000}
azure:
+ - name: Claude Haiku 5.5
+ model: claude-haiku-5-5
+ description: "Small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-10-07
+ cost: {input: 0.1, output: 0.5}
+ reasoning_options:
+ - type: toggle
+ - type: effort
+ values: [low, medium, high, xhigh, max]
+ limits: {context: 1000000, output: 128000}
+
- name: GPT-6.1 Sol
model: gpt-6.1-sol
description: "Reasoning model for agentic coding, computer use, and document-heavy professional work"
@@ -846,28 +908,6 @@ azure:
values: [none, low, medium, high, xhigh, max]
limits: {context: 1050000, output: 128000}
- - 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}
- limits: {context: 0, output: 0}
-
- - 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}
- limits: {context: 0, output: 0}
-
- name: GPT-6 Astra
model: gpt-6-astra
description: "OpenAI model for advanced analysis, software engineering, deep research, scientific work, and document creation"
@@ -908,6 +948,7 @@ azure:
- type: effort
values: [low, medium, high, xhigh]
limits: {context: 200000, output: 128000}
+ cost: {input: 2, output: 6}
- name: Claude Opus 5
model: claude-opus-5
@@ -1100,6 +1141,34 @@ azure:
limits: {context: 1000000, output: 384000}
ollama:
+ - name: Muse Spark Open
+ model: muse-spark-open
+ description: "Open-weight model optimized for agentic coding, reliable tool calling, and codebase understanding"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image]
+ output_types: [text]
+ release_date: 2026-10-15
+ cost: {input: 1.25, output: 4.25}
+ reasoning_options:
+ - type: effort
+ values: [low, medium, high, xhigh]
+ limits: {context: 1048576, output: 131072}
+
+ - name: Mistral Large 4
+ model: mistral-large-4
+ description: "Multimodal model from Mistral AI for reasoning, coding, and agentic workloads"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image]
+ output_types: [text]
+ release_date: 2026-10-06
+ cost: {input: 1.36, output: 4.18}
+ reasoning_options:
+ - type: effort
+ values: [none, high]
+ limits: {context: 1048576, output: 262144}
+
- name: DeepSeek V4.1 Flash
model: deepseek-v4.1-flash
description: "Sparse Mixture-of-Experts model from DeepSeek built on a Causal Encoder-Decoder architecture"
@@ -1313,6 +1382,19 @@ ollama:
cost: {input: 0.6, output: 3.6}
wandb:
+ - name: Nemotron 3.5 Super VL
+ model: nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B
+ description: "Open multimodal Mixture-of-Experts model with 12B active parameters out of 120B for reasoning, coding, tool use, and document understanding"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image]
+ output_types: [text]
+ release_date: 2026-10-15
+ cost: {input: 0.15, output: 0.75}
+ reasoning_options:
+ - type: toggle
+ limits: {context: 1048576, output: 1048576}
+
- 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"
@@ -1321,7 +1403,7 @@ wandb:
input_types: [text, image]
output_types: [text]
release_date: 2026-09-10
- cost: {input: 0.2, output: 0.65}
+ cost: {input: 0.15, output: 0.65}
reasoning_options:
- type: toggle
limits: {context: 1048576, output: 1048576}
@@ -1449,7 +1531,7 @@ wandb:
input_types: [text]
output_types: [text]
release_date: 2026-06-16
- cost: {input: 0.76, output: 2.42}
+ cost: {input: 0.62, output: 2.42}
reasoning_options:
- type: toggle
limits: {context: 1048576, output: 1048576}
@@ -1513,6 +1595,35 @@ wandb:
limits: {context: 131072, output: 131072}
opencode-go:
+ - name: Gemini 4 Argon Contributor
+ model: gemini-4-argon-contributor
+ description: "Contributor tier of Gemini 4 Argon for reasoning, coding, and agents, with prompts and completions eligible for model training"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-10-13
+ cost: {input: 4, output: 20}
+ reasoning_options:
+ - type: effort
+ values: [low, medium, high]
+ limits: {context: 1048576, output: 65536}
+
+ - name: Claude Haiku 5.5
+ model: claude-haiku-5-5
+ description: "Small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-10-07
+ cost: {input: 0.1, output: 0.5}
+ reasoning_options:
+ - type: toggle
+ - type: effort
+ values: [low, medium, high, xhigh, max]
+ limits: {context: 1000000, output: 128000}
+
- 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"