models.yml

anthropic:

  - name: Claude Fable 5
    model: claude-fable-5
    description: "Anthropic's most capable widely released Mythos-class model, built for the most demanding reasoning and long-horizon agentic work"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-06-09
    cost: {input: 10, output: 50}

  - name: Claude Sonnet 5
    model: 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 Opus 4.8
    model: claude-opus-4-8
    description: "Modest but tangible improvement over Claude Opus 4.7"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-05-28
    cost: {input: 5, output: 25}

  - name: Claude Opus 4.7
    model: claude-opus-4-7
    description: "Next generation of Anthropic's Opus family, 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 4.6
    model: claude-sonnet-4-6
    description: "Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-02-17
    cost: {input: 3, output: 15}

  - name: Claude Opus 4.6
    model: claude-opus-4-6
    description: "Anthropic's strongest model for coding and long-running professional tasks"
    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}

  - name: Claude Opus 4.5
    model: claude-opus-4-5
    description: "Anthropic's frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use"
    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}

  - 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: 2025-10-15
    cost: {input: 1, output: 5}

  - name: Claude Sonnet 4.5
    model: claude-sonnet-4-5
    description: "Anthropic's most advanced Sonnet model to date, optimized for real-world agents and coding workflows"
    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}

openai:
  - name: GPT-Realtime-2.1
    model: gpt-realtime-2.1
    description: ""
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-07-06
    cost: {input: 4, output: 24}

  - 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"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-04-23
    cost: {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's latest frontier model, unifying 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 Chat
    model: gpt-5.3-chat-latest
    description: "Update to ChatGPT's most-used model that makes everyday conversations smoother, more useful, and more directly helpful"
    roles: [chat, edit]
    capabilities: [tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-03-03
    cost: {input: 1.75, output: 14}

  - name: GPT-5.3 Codex
    model: gpt-5.3-codex
    description: "OpenAI's most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2"
    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

google:
  - name: Claude Sonnet 5
    model: claude-sonnet-5@default
    description: ""
    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: Gemini Omni Flash Preview
    model: gemini-omni-flash-preview
    description: ""
    roles: [chat, edit]
    capabilities: [thinking]
    input_types: [text, image]
    output_types: []
    release_date: 2026-06-30
    cost: {input: 1.5, output: 17.5}

  - name: Claude Opus 4.8
    model: claude-opus-4-8@default
    description: ""
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-05-28
    cost: {input: 5, output: 25}

  - name: Nano Banana 2
    model: gemini-3.1-flash-image
    description: ""
    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}

  - 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 IT
    model: gemma-4-26b-a4b-it
    description: "Instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-04-02

  - name: Gemma 4 31B IT
    model: gemma-4-31b-it
    description: "Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-04-02

  - name: Nano Banana 2
    model: gemini-3.1-flash-image-preview
    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: 2026-02-26
    cost: {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}

bedrock:
  - name: GPT-5.6 Luna
    model: openai.gpt-5.6-luna
    description: ""
    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: openai.gpt-5.6-sol
    description: ""
    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: openai.gpt-5.6-terra
    description: ""
    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: Claude Sonnet 5 (AU)
    model: au.anthropic.claude-sonnet-5
    description: ""
    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 Fable 5 (Global)
    model: global.anthropic.claude-fable-5
    description: "Anthropic's most capable widely released Mythos-class model, built for the most demanding reasoning and long-horizon agentic work"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    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"
    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 Opus 4.8 (Global)
    model: global.anthropic.claude-opus-4-8
    description: "Modest but tangible improvement over Claude Opus 4.7"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-05-28
    cost: {input: 5, output: 25}

  - name: Claude Opus 4.7
    model: anthropic.claude-opus-4-7
    description: "Next generation of Anthropic's Opus family, 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: "Next generation of Anthropic's Opus family, 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: "SOTA 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-5
    model: zai.glm-5
    description: "Z.ai's flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-03-18
    cost: {input: 1, output: 3.2}

  - name: NVIDIA Nemotron 3 Super 120B A12B
    model: nvidia.nemotron-super-3-120b
    description: "120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-03-11
    cost: {input: 0.15, output: 0.65}

  - name: Claude Sonnet 4.6
    model: anthropic.claude-sonnet-4-6
    description: "Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-02-17
    cost: {input: 3, output: 15}

azure:
  - name: GPT-5.6 Luna
    model: gpt-5.6-luna
    description: ""
    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: ""
    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: ""
    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: Claude Fable 5
    model: claude-fable-5
    description: "Anthropic's most capable widely released Mythos-class model, built for the most demanding reasoning and long-horizon agentic work"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-06-09
    cost: {input: 10, output: 50}

  - name: Claude Sonnet 5
    model: 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: 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"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-04-24
    cost: {input: 5, output: 30}

  - name: Kimi K2.6
    model: kimi-k2.6
    description: "Moonshot AI's next-generation multimodal model, designed 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-reasoning
    description: "Reasoning model from xAI with industry-leading speed and agentic tool calling capabilities"
    roles: [chat, edit]
    capabilities: [tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-08
    cost: {input: 2, output: 6}

  - name: Grok 4.20 (Reasoning)
    model: grok-4-20-reasoning
    description: "Reasoning model from xAI with industry-leading speed and agentic tool calling capabilities"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-08
    cost: {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's latest frontier model, unifying 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 Chat
    model: gpt-5.3-chat
    description: "Update to ChatGPT's most-used model that makes everyday conversations smoother, more useful, and more directly helpful"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-03-03
    cost: {input: 1.75, output: 14}

  - name: GPT-5.3 Codex
    model: gpt-5.3-codex
    description: "OpenAI's most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2"
    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 4.8
    model: claude-opus-4-8
    description: "Modest but tangible improvement over Claude Opus 4.7"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-05-28
    cost: {input: 5, output: 25}

  - name: Claude Sonnet 4.6
    model: claude-sonnet-4-6
    description: "Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image, pdf]
    output_types: [text]
    release_date: 2026-02-17
    cost: {input: 3, output: 15}

  - name: DeepSeek-V4-Flash
    model: deepseek-v4-flash
    description: "Efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window"
    roles: [chat, edit]
    capabilities: [thinking]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-24
    cost: {input: 0.19, output: 0.51}

  - name: DeepSeek-V4-Pro
    model: deepseek-v4-pro
    description: "Large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window"
    roles: [chat, edit]
    capabilities: [thinking]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-24
    cost: {input: 1.74, output: 3.48}

github:
  - name: DeepSeek-R1-0528
    model: deepseek/deepseek-r1-0528
    description: "May 28th update to DeepSeek R1, with performance on par with OpenAI o1 but open-sourced and with fully open reasoning tokens"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2025-05-28
    cost: {input: 0, output: 0}

  - name: Mistral Medium 3 (25.05)
    model: mistral-ai/mistral-medium-2505
    description: "High-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-05-01
    cost: {input: 0, output: 0}

  - name: GPT-4.1
    model: openai/gpt-4.1
    description: "Flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning"
    roles: [chat, edit]
    capabilities: [tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-04-14
    cost: {input: 0, output: 0}

  - name: GPT-4.1-mini
    model: openai/gpt-4.1-mini
    description: "Mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost"
    roles: [chat, edit]
    capabilities: [tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-04-14
    cost: {input: 0, output: 0}

  - name: GPT-4.1-nano
    model: openai/gpt-4.1-nano
    description: "For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series"
    roles: [chat, edit]
    capabilities: [tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-04-14
    cost: {input: 0, output: 0}

  - name: DeepSeek-V3-0324
    model: deepseek/deepseek-v3-0324
    description: "Latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2025-03-24
    cost: {input: 0, output: 0}

  - name: Mistral Small 3.1
    model: mistral-ai/mistral-small-2503
    description: "Upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-03-01
    cost: {input: 0, output: 0}

  - name: Llama 4 Maverick 17B 128E Instruct FP8
    model: meta/llama-4-maverick-17b-128e-instruct-fp8
    description: "High-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-01-31
    cost: {input: 0, output: 0}

  - name: Llama 4 Scout 17B 16E Instruct
    model: meta/llama-4-scout-17b-16e-instruct
    description: "Mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-01-31
    cost: {input: 0, output: 0}

  - name: OpenAI o3
    model: openai/o3
    description: "Well-rounded reasoning model setting a new standard for math, science, coding, and visual reasoning tasks"
    roles: [chat, edit]
    capabilities: [thinking]
    input_types: [text, image]
    output_types: [text]
    release_date: 2025-01-31
    cost: {input: 0, output: 0}

ollama:
  - name: DeepSeek V4 Flash
    model: deepseek-v4-flash
    description: "Efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-24

  - name: DeepSeek V4 Pro
    model: deepseek-v4-pro
    description: "Large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    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's next-generation multimodal model, designed 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 31B
    model: gemma4:31b
    description: "Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-04-02

  - name: GLM 5.2
    model: glm-5.2
    description: "Large-scale reasoning model from Z.ai with a 1M-token context window, suited for long-horizon agent workflows and project-level software engineering"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    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 M3
    model: minimax-m3
    description: "Multimodal foundation model from MiniMax with a 1M-token context window, suited for long-horizon agentic work and coding"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-05-31

  - name: MiniMax M2.7
    model: minimax-m2.7
    description: "Next-generation large language model designed for autonomous, real-world productivity and continuous improvement"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-03-18

  - name: NVIDIA Nemotron 3 Ultra
    model: nemotron-3-ultra
    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

  - name: NVIDIA Nemotron 3 Super 120B
    model: nemotron-3-super
    description: "120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-03-11

  - name: Qwen 3.5 397B
    model: qwen3.5:397b
    description: "Built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    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"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-24
    cost: {input: 0.14, output: 0.28}

  - name: DeepSeek V4-Pro
    model: deepseek-ai/DeepSeek-V4-Pro
    description: "DeepSeek V4-Pro is a 1.6T-parameter MoE model with 49B active parameters excelling at advanced reasoning, coding, and complex agentic workloads"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-24
    cost: {input: 1.74, output: 3.46}

  - name: Z.AI GLM 5.2
    model: zai-org/GLM-5.2
    description: "GLM-5.2 is a Mixture-of-Experts language model featuring 40 billion activated parameters and a total of 744 billion parameters"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-06-13
    cost: {input: 1.39, output: 4.4}

  - 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 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"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-03-11
    cost: {input: 0.2, output: 0.8}

  - name: MiniMax M2.5
    model: MiniMaxAI/MiniMax-M2.5
    description: "SOTA large language model designed for real-world productivity"
    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}

  - name: Moonshot AI 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"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-06-12
    cost: {input: 0.94, output: 4}

  - 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"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-04-20
    cost: {input: 0.95, output: 4}

  - 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: 2025-08-05
    cost: {input: 0.04, output: 0.14}

  - 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: 2025-08-05
    cost: {input: 0.03, output: 0.13}

  - 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"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-04-16
    cost: {input: 0.25, output: 1.25}

  - 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"
    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}

  - name: JetBrains 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]
    output_types: [text]
    release_date: 2026-06-01
    cost: {input: 0.05, output: 0.1}

opencode-go:
  - name: Kimi K3 (2x usage)
    model: kimi-k3
    description: ""
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-07-16
    cost: {input: 3, output: 15}

  - name: Grok 4.5
    model: grok-4.5
    description: ""
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-07-08
    cost: {input: 2, output: 6}

  - name: DeepSeek V4 Flash
    model: deepseek-v4-flash
    description: "Efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-24
    cost: {input: 0.14, output: 0.28}

  - name: DeepSeek V4 Pro
    model: deepseek-v4-pro
    description: "Large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-04-24
    cost: {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's next-generation multimodal model, designed 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 5.2
    model: glm-5.2
    description: "Large-scale reasoning model from Z.ai with a 1M-token context window, suited for long-horizon agent workflows and project-level software engineering"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-06-13
    cost: {input: 1.4, output: 4.4}

  - 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: 2026-04-07
    cost: {input: 1.4, output: 4.4}

  - name: Qwen3.7 Plus
    model: qwen3.7-plus
    description: "Cost-effective model in Alibaba's Qwen3.7 series supporting text and image input with upgraded text capabilities"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-06-02
    cost: {input: 0.4, output: 1.6}

  - name: Qwen3.7 Max
    model: qwen3.7-max
    description: "Capable of writing and debugging code, and sustaining autonomous execution across hundreds or thousands of steps"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text]
    output_types: [text]
    release_date: 2026-05-21
    cost: {input: 2.5, output: 7.5}

  - name: Qwen3.6 Plus
    model: qwen3.6-plus
    description: "Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-04-02
    cost: {input: 0.5, output: 3}

  - name: MiniMax M3 (3x usage)
    model: minimax-m3
    description: "Multimodal foundation model from MiniMax with a 1M-token context window, suited for long-horizon agentic work and coding"
    roles: [chat, edit]
    capabilities: [thinking, tool_calling]
    input_types: [text, image]
    output_types: [text]
    release_date: 2026-05-31
    cost: {input: 0.1, output: 0.4}

  - name: MiniMax M2.7
    model: minimax-m2.7
    description: "Next-generation large language model designed for autonomous, real-world productivity and continuous improvement"
    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: 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}