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 (#10157)

github-actions[bot]959bddbde555
diff --git a/packages/llm-info/data/models.yml b/packages/llm-info/data/models.yml
index 2c84fc0533..0a33127f64 100644
--- a/packages/llm-info/data/models.yml
+++ b/packages/llm-info/data/models.yml
@@ -91,6 +91,46 @@ anthropic:
cost: {input: 3, output: 15}
openai:
+ - name: GPT-5.6
+ model: gpt-5.6
+ description: "OpenAI's GPT-5.6 flagship tier for frontier reasoning, coding, and agentic workflows"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-07-09
+ cost: {input: 5, output: 30}
+
+ - name: GPT-5.6 Luna
+ model: gpt-5.6-luna
+ description: "Fast, cost-efficient model in OpenAI's GPT-5.6 series, suited for high-volume, latency-sensitive chat and lightweight agentic workflows"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-07-09
+ cost: {input: 1, output: 6}
+
+ - name: GPT-5.6 Sol
+ model: gpt-5.6-sol
+ description: "Flagship model in OpenAI's GPT-5.6 series, suited for complex reasoning, coding, and agentic workflows"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-07-09
+ cost: {input: 5, output: 30}
+
+ - name: GPT-5.6 Terra
+ model: gpt-5.6-terra
+ description: "Balanced model in OpenAI's GPT-5.6 series, suited for everyday coding, reasoning, and agentic workflows"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-07-09
+ cost: {input: 2.5, output: 15}
+
- name: GPT-5.5
model: gpt-5.5
description: "OpenAI's frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks"
@@ -240,6 +280,16 @@ bedrock:
release_date: 2026-06-09
cost: {input: 10, output: 50}
+ - name: Claude Sonnet 5 (Global)
+ model: global.anthropic.claude-sonnet-5
+ description: "Anthropic's most agentic Sonnet model yet, built for planning, tool use, and autonomous multi-step work at Sonnet pricing"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text, image, pdf]
+ output_types: [text]
+ release_date: 2026-06-30
+ cost: {input: 2, output: 10}
+
- name: Claude Sonnet 5
model: anthropic.claude-sonnet-5
description: "Anthropic's most agentic Sonnet model yet, built for planning, tool use, and autonomous multi-step work at Sonnet pricing"
@@ -682,6 +732,16 @@ ollama:
release_date: 2026-02-15
wandb:
+ - 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-Flash
model: deepseek-ai/DeepSeek-V4-Flash
description: "DeepSeek V4-Flash is an MoE model with 1M context length great for coding, reasoning, and agentic workloads"
@@ -722,6 +782,16 @@ wandb:
release_date: 2026-04-07
cost: {input: 1.4, output: 4.4}
+ - name: Nemotron 3 Ultra
+ model: nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B
+ description: "Open frontier-reasoning and orchestration model from NVIDIA with 55B active parameters out of 550B total (MoE)"
+ roles: [chat, edit]
+ capabilities: [thinking, tool_calling]
+ input_types: [text]
+ output_types: [text]
+ release_date: 2026-06-04
+ cost: {input: 0.75, output: 2.75}
+
- name: NVIDIA Nemotron 3 Super 120B
model: nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8
description: "120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications"