models.json

{
  "models": [
    {
      "id": "gpt-5",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-5",
      "multiModal": true
    },
    {
      "id": "gpt-5-mini",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-5 Mini",
      "multiModal": true
    },
    {
      "id": "gpt-5-nano",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-5 Nano",
      "multiModal": true
    },
    {
      "id": "o3",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "o3",
      "multiModal": true
    },
    {
      "id": "o3-mini",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "o3-mini",
      "multiModal": false
    },
    {
      "id": "o4-mini",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "o4-mini",
      "multiModal": true
    },
    {
      "id": "gpt-4.1",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-4.1",
      "multiModal": true
    },
    {
      "id": "gpt-4.1-mini",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-4.1 mini",
      "multiModal": true
    },
    {
      "id": "gpt-4.1-nano",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-4.1 nano",
      "multiModal": true
    },
    {
      "id": "gpt-4.5-preview",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-4.5 (preview)",
      "multiModal": true
    },
    {
      "id": "gpt-4o",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-4o",
      "multiModal": true
    },
    {
      "id": "gpt-4o-mini",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-4o mini",
      "multiModal": true
    },
    {
      "id": "o1",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "o1",
      "multiModal": true
    },
    {
      "id": "o1-mini",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "o1-mini",
      "multiModal": false
    },
    {
      "id": "gpt-4-turbo",
      "provider": "OpenAI",
      "providerId": "openai",
      "name": "GPT-4 Turbo",
      "multiModal": true
    },
    {
      "id": "claude-sonnet-4-5-20250929",
      "provider": "Anthropic",
      "providerId": "anthropic",
      "name": "Claude Sonnet 4.5",
      "multiModal": true
    },
    {
      "id": "claude-opus-4-1-20250805",
      "provider": "Anthropic",
      "providerId": "anthropic",
      "name": "Claude Opus 4.1",
      "multiModal": true
    },
    {
      "id": "claude-opus-4-20250514",
      "provider": "Anthropic",
      "providerId": "anthropic",
      "name": "Claude Opus 4",
      "multiModal": true
    },
    {
      "id": "claude-sonnet-4-20250514",
      "provider": "Anthropic",
      "providerId": "anthropic",
      "name": "Claude Sonnet 4",
      "multiModal": true
    },
    {
      "id": "claude-3-7-sonnet-latest",
      "provider": "Anthropic",
      "providerId": "anthropic",
      "name": "Claude Sonnet 3.7",
      "multiModal": true
    },
    {
      "id": "claude-3-5-sonnet-latest",
      "provider": "Anthropic",
      "providerId": "anthropic",
      "name": "Claude Sonnet 3.5",
      "multiModal": true
    },
    {
      "id": "claude-3-5-haiku-latest",
      "provider": "Anthropic",
      "providerId": "anthropic",
      "name": "Claude Haiku 3.5",
      "multiModal": true
    },
    {
      "id": "gemini-2.5-flash-preview-05-20",
      "provider": "Google Vertex AI",
      "providerId": "vertex",
      "name": "Gemini 2.5 Flash Preview 05-20",
      "multiModal": true
    },
    {
      "id": "gemini-2.5-pro-preview-05-06",
      "provider": "Google Vertex AI",
      "providerId": "vertex",
      "name": "Gemini 2.5 Pro Preview 05-06",
      "multiModal": true
    },
    {
      "id": "gemini-2.0-flash-001",
      "provider": "Google Vertex AI",
      "providerId": "vertex",
      "name": "Gemini 2.0 Flash",
      "multiModal": true
    },
    {
      "id": "gemini-2.0-flash-lite-001",
      "provider": "Google Vertex AI",
      "providerId": "vertex",
      "name": "Gemini 2.0 Flash Lite",
      "multiModal": true
    },
    {
      "id": "gemini-1.5-pro-002",
      "provider": "Google Vertex AI",
      "providerId": "vertex",
      "name": "Gemini 1.5 Pro",
      "multiModal": true
    },
    {
      "id": "gemini-1.5-flash-002",
      "provider": "Google Vertex AI",
      "providerId": "vertex",
      "name": "Gemini 1.5 Flash",
      "multiModal": true
    },
    {
      "id": "models/gemini-2.5-flash-preview-05-20",
      "provider": "Google Generative AI",
      "providerId": "google",
      "name": "Gemini 2.5 Flash Preview 05-20",
      "multiModal": true
    },
    {
      "id": "models/gemini-2.5-pro-preview-05-06",
      "provider": "Google Generative AI",
      "providerId": "google",
      "name": "Gemini 2.5 Pro Preview 05-06",
      "multiModal": true
    },
    {
      "id": "models/gemini-2.0-flash",
      "provider": "Google Generative AI",
      "providerId": "google",
      "name": "Gemini 2.0 Flash",
      "multiModal": true
    },
    {
      "id": "models/gemini-2.0-flash-lite",
      "provider": "Google Generative AI",
      "providerId": "google",
      "name": "Gemini 2.0 Flash Lite",
      "multiModal": true
    },
    {
      "id": "models/gemini-1.5-pro",
      "provider": "Google Generative AI",
      "providerId": "google",
      "name": "Gemini 1.5 Pro",
      "multiModal": true
    },
    {
      "id": "models/gemini-1.5-flash",
      "provider": "Google Generative AI",
      "providerId": "google",
      "name": "Gemini 1.5 Flash",
      "multiModal": true
    },
    {
      "id": "magistral-medium-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Magistral Medium",
      "multiModal": true
    },
    {
      "id": "magistral-small-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Magistral Small",
      "multiModal": true
    },
    {
      "id": "mistral-medium-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Mistral Medium",
      "multiModal": true
    },
    {
      "id": "mistral-large-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Mistral Large",
      "multiModal": false
    },
    {
      "id": "pixtral-large-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Pixtral Large",
      "multiModal": true
    },
    {
      "id": "mistral-small-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Mistral Small",
      "multiModal": false
    },
    {
      "id": "open-mistral-nemo",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Mistral Nemo",
      "multiModal": false
    },
    {
      "id": "devstral-small-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Devstral Small",
      "multiModal": false
    },
    {
      "id": "codestral-latest",
      "provider": "Mistral",
      "providerId": "mistral",
      "name": "Codestral 25.01",
      "multiModal": false
    },
    {
      "id": "openai/gpt-oss-120b",
      "provider": "Groq",
      "providerId": "groq",
      "name": "GPT OSS 120B",
      "multiModal": false
    },
    {
      "id": "moonshotai/kimi-k2-instruct",
      "provider": "Groq",
      "providerId": "groq",
      "name": "Kimi K2 Instruct",
      "multiModal": false
    },
    {
      "id": "qwen/qwen3-32b",
      "provider": "Groq",
      "providerId": "groq",
      "name": "Qwen3 32B",
      "multiModal": false
    },
    {
      "id": "llama-3.3-70b-versatile",
      "provider": "Groq",
      "providerId": "groq",
      "name": "Llama 3.3 70B",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/qwen3-coder-480b-a35b-instruct",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Qwen3 Coder 480B A35B Instruct",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/qwen3-235b-a22b-thinking-2507",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Qwen3 235B-A22B-Thinking-2507",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/qwen3-235b-a22b-instruct-2507",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Qwen3 235B-A22B-Instruct-2507",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/zai-org/glm-4p5",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Z.ai GLM 4.5",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/kimi-k2-instruct",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Kimi K2 Instruct",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/deepseek-r1",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "DeepSeek R1",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/llama4-maverick-instruct-basic",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Llama 4 Maverick Instruct",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/llama4-scout-instruct-basic",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Llama 4 Scout Instruct",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/llama-v3p1-405b-instruct",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Llama 3.1 405B",
      "multiModal": false
    },
    {
      "id": "accounts/fireworks/models/llama-v3p1-70b-instruct",
      "provider": "Fireworks",
      "providerId": "fireworks",
      "name": "Llama 3.1 70B",
      "multiModal": false
    },
    {
      "id": "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo",
      "provider": "Together AI",
      "providerId": "togetherai",
      "name": "Llama 3.1 70B",
      "multiModal": false
    },
    {
      "id": "grok-4",
      "provider": "xAI",
      "providerId": "xai",
      "name": "Grok 4",
      "multiModal": true
    },
    {
      "id": "grok-3",
      "provider": "xAI",
      "providerId": "xai",
      "name": "Grok 3",
      "multiModal": false
    },
    {
      "id": "grok-3-mini",
      "provider": "xAI",
      "providerId": "xai",
      "name": "Grok 3 Mini",
      "multiModal": false
    },
    {
      "id": "grok-3-fast",
      "provider": "xAI",
      "providerId": "xai",
      "name": "Grok 3 Fast",
      "multiModal": false
    },
    {
      "id": "grok-3-mini-fast",
      "provider": "xAI",
      "providerId": "xai",
      "name": "Grok 3 Mini Fast",
      "multiModal": false
    },
    {
      "id": "grok-code-fast-1",
      "provider": "xAI",
      "providerId": "xai",
      "name": "Grok Code Fast 1",
      "multiModal": false
    },
    {
      "id": "deepseek-chat",
      "provider": "DeepSeek",
      "providerId": "deepseek",
      "name": "DeepSeek V3",
      "multiModal": false
    },
    {
      "id": "llama3.1",
      "provider": "Ollama",
      "providerId": "ollama",
      "name": "Llama 3.1",
      "multiModal": false
    },
    {
      "id": "mistral-nemo",
      "provider": "Ollama",
      "providerId": "ollama",
      "name": "Mistral Nemo",
      "multiModal": false
    },
    {
      "id": "mistral-large",
      "provider": "Ollama",
      "providerId": "ollama",
      "name": "Mistral Large",
      "multiModal": false
    },
    {
      "id": "mistralai/voxtral-small-24b-2507",
      "provider": "Mistral: Voxtral Small 24B 2507",
      "providerId": "openrouter",
      "name": "Mistral: Voxtral Small 24B 2507",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000003",
        "request": "0",
        "image": "0",
        "audio": "0.0001",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32000,
      "description": "Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio is priced at $100 per million seconds."
    },
    {
      "id": "openai/gpt-oss-safeguard-20b",
      "provider": "OpenAI: gpt-oss-safeguard-20b",
      "providerId": "openrouter",
      "name": "OpenAI: gpt-oss-safeguard-20b",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000000075",
        "completion": "0.0000003",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000037"
      },
      "contextLength": 131072,
      "description": "gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust & safety labeling.\n\nLearn more about this model in OpenAI's gpt-oss-safeguard [user guide](https://cookbook.openai.com/articles/gpt-oss-safeguard-guide)."
    },
    {
      "id": "qwen/qwen3-vl-32b-instruct",
      "provider": "Qwen: Qwen3 VL 32B Instruct",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 VL 32B Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000035",
        "completion": "0.0000011",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video. With 32 billion parameters, it combines deep visual perception with advanced text comprehension, enabling fine-grained spatial reasoning, document and scene analysis, and long-horizon video understanding.Robust OCR in 32 languages, and enhanced multimodal fusion through Interleaved-MRoPE and DeepStack architectures. Optimized for agentic interaction and visual tool use, Qwen3-VL-32B delivers state-of-the-art performance for complex real-world multimodal tasks."
    },
    {
      "id": "openai/gpt-5-image-mini",
      "provider": "OpenAI: GPT-5 Image Mini",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5 Image Mini",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000025",
        "completion": "0.000002",
        "request": "0",
        "image": "0.0000025",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000025"
      },
      "contextLength": 400000,
      "description": "GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation. This natively multimodal model features superior instruction following, text rendering, and detailed image editing with reduced latency and cost. It excels at high-quality visual creation while maintaining strong text understanding, making it ideal for applications that require both efficient image generation and text processing at scale."
    },
    {
      "id": "anthropic/claude-haiku-4.5",
      "provider": "Anthropic: Claude Haiku 4.5",
      "providerId": "openrouter",
      "name": "Anthropic: Claude Haiku 4.5",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000001",
        "completion": "0.000005",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000001",
        "input_cache_write": "0.00000125"
      },
      "contextLength": 200000,
      "description": "Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance across reasoning, coding, and computer-use tasks, Haiku 4.5 brings frontier-level capability to real-time and high-volume applications.\n\nIt introduces extended thinking to the Haiku line; enabling controllable reasoning depth, summarized or interleaved thought output, and tool-assisted workflows with full support for coding, bash, web search, and computer-use tools. Scoring >73% on SWE-bench Verified, Haiku 4.5 ranks among the world’s best coding models while maintaining exceptional responsiveness for sub-agents, parallelized execution, and scaled deployment."
    },
    {
      "id": "qwen/qwen3-vl-8b-thinking",
      "provider": "Qwen: Qwen3 VL 8B Thinking",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 VL 8B Thinking",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000018",
        "completion": "0.0000021",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 256000,
      "description": "Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences. It integrates enhanced multimodal alignment and long-context processing (native 256K, expandable to 1M tokens) for tasks such as scientific visual analysis, causal inference, and mathematical reasoning over image or video inputs.\n\nCompared to the Instruct edition, the Thinking version introduces deeper visual-language fusion and deliberate reasoning pathways that improve performance on long-chain logic tasks, STEM problem-solving, and multi-step video understanding. It achieves stronger temporal grounding via Interleaved-MRoPE and timestamp-aware embeddings, while maintaining robust OCR, multilingual comprehension, and text generation on par with large text-only LLMs."
    },
    {
      "id": "qwen/qwen3-vl-8b-instruct",
      "provider": "Qwen: Qwen3 VL 8B Instruct",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 VL 8B Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000008",
        "completion": "0.0000005",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0"
      },
      "contextLength": 131072,
      "description": "Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video. It features improved multimodal fusion with Interleaved-MRoPE for long-horizon temporal reasoning, DeepStack for fine-grained visual-text alignment, and text-timestamp alignment for precise event localization.\n\nThe model supports a native 256K-token context window, extensible to 1M tokens, and handles both static and dynamic media inputs for tasks like document parsing, visual question answering, spatial reasoning, and GUI control. It achieves text understanding comparable to leading LLMs while expanding OCR coverage to 32 languages and enhancing robustness under varied visual conditions."
    },
    {
      "id": "openai/gpt-5-image",
      "provider": "OpenAI: GPT-5 Image",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5 Image",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00001",
        "completion": "0.00001",
        "request": "0",
        "image": "0.00001",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000125"
      },
      "contextLength": 400000,
      "description": "[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's most advanced language model with state-of-the-art image generation capabilities. It offers major improvements in reasoning, code quality, and user experience while incorporating GPT Image 1's superior instruction following, text rendering, and detailed image editing."
    },
    {
      "id": "openai/o3-deep-research",
      "provider": "OpenAI: o3 Deep Research",
      "providerId": "openrouter",
      "name": "OpenAI: o3 Deep Research",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00001",
        "completion": "0.00004",
        "request": "0",
        "image": "0.00765",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000025"
      },
      "contextLength": 200000,
      "description": "o3-deep-research is OpenAI's advanced model for deep research, designed to tackle complex, multi-step research tasks.\n\nNote: This model always uses the 'web_search' tool which adds additional cost."
    },
    {
      "id": "openai/o4-mini-deep-research",
      "provider": "OpenAI: o4 Mini Deep Research",
      "providerId": "openrouter",
      "name": "OpenAI: o4 Mini Deep Research",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000008",
        "request": "0",
        "image": "0.00153",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000005"
      },
      "contextLength": 200000,
      "description": "o4-mini-deep-research is OpenAI's faster, more affordable deep research model—ideal for tackling complex, multi-step research tasks.\n\nNote: This model always uses the 'web_search' tool which adds additional cost."
    },
    {
      "id": "google/gemini-2.5-flash-image",
      "provider": "Google: Gemini 2.5 Flash Image (Nano Banana)",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Flash Image (Nano Banana)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000025",
        "request": "0",
        "image": "0.001238",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Gemini 2.5 Flash Image, a.k.a. \"Nano Banana,\" is now generally available. It is a state of the art image generation model with contextual understanding. It is capable of image generation, edits, and multi-turn conversations. Aspect ratios can be controlled with the [image_config API Parameter](https://openrouter.ai/docs/features/multimodal/image-generation#image-aspect-ratio-configuration)"
    },
    {
      "id": "qwen/qwen3-vl-30b-a3b-thinking",
      "provider": "Qwen: Qwen3 VL 30B A3B Thinking",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 VL 30B A3B Thinking",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000002",
        "completion": "0.000001",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0"
      },
      "contextLength": 131072,
      "description": "Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Thinking variant enhances reasoning in STEM, math, and complex tasks. It excels in perception of real-world/synthetic categories, 2D/3D spatial grounding, and long-form visual comprehension, achieving competitive multimodal benchmark results. For agentic use, it handles multi-image multi-turn instructions, video timeline alignments, GUI automation, and visual coding from sketches to debugged UI. Text performance matches flagship Qwen3 models, suiting document AI, OCR, UI assistance, spatial tasks, and agent research."
    },
    {
      "id": "qwen/qwen3-vl-30b-a3b-instruct",
      "provider": "Qwen: Qwen3 VL 30B A3B Instruct",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 VL 30B A3B Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000015",
        "completion": "0.0000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Instruct variant optimizes instruction-following for general multimodal tasks. It excels in perception of real-world/synthetic categories, 2D/3D spatial grounding, and long-form visual comprehension, achieving competitive multimodal benchmark results. For agentic use, it handles multi-image multi-turn instructions, video timeline alignments, GUI automation, and visual coding from sketches to debugged UI. Text performance matches flagship Qwen3 models, suiting document AI, OCR, UI assistance, spatial tasks, and agent research."
    },
    {
      "id": "openai/gpt-5-pro",
      "provider": "OpenAI: GPT-5 Pro",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5 Pro",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000015",
        "completion": "0.00012",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0"
      },
      "contextLength": 400000,
      "description": "GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. It supports test-time routing features and advanced prompt understanding, including user-specified intent like \"think hard about this.\" Improvements include reductions in hallucination, sycophancy, and better performance in coding, writing, and health-related tasks."
    },
    {
      "id": "anthropic/claude-sonnet-4.5",
      "provider": "Anthropic: Claude Sonnet 4.5",
      "providerId": "openrouter",
      "name": "Anthropic: Claude Sonnet 4.5",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000003",
        "completion": "0.000015",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 1000000,
      "description": "Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with improvements across system design, code security, and specification adherence. The model is designed for extended autonomous operation, maintaining task continuity across sessions and providing fact-based progress tracking.\n\nSonnet 4.5 also introduces stronger agentic capabilities, including improved tool orchestration, speculative parallel execution, and more efficient context and memory management. With enhanced context tracking and awareness of token usage across tool calls, it is particularly well-suited for multi-context and long-running workflows. Use cases span software engineering, cybersecurity, financial analysis, research agents, and other domains requiring sustained reasoning and tool use."
    },
    {
      "id": "google/gemini-2.5-flash-preview-09-2025",
      "provider": "Google: Gemini 2.5 Flash Preview 09-2025",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Flash Preview 09-2025",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000025",
        "request": "0",
        "image": "0.001238",
        "audio": "0.000001",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000075",
        "input_cache_write": "0.0000003833"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Flash Preview September 2025 Checkpoint is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in \"thinking\" capabilities, enabling it to provide responses with greater accuracy and nuanced context handling. \n\nAdditionally, Gemini 2.5 Flash is configurable through the \"max tokens for reasoning\" parameter, as described in the documentation (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
    },
    {
      "id": "google/gemini-2.5-flash-lite-preview-09-2025",
      "provider": "Google: Gemini 2.5 Flash Lite Preview 09-2025",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Flash Lite Preview 09-2025",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance across common benchmarks compared to earlier Flash models. By default, \"thinking\" (i.e. multi-pass reasoning) is disabled to prioritize speed, but developers can enable it via the [Reasoning API parameter](https://openrouter.ai/docs/use-cases/reasoning-tokens) to selectively trade off cost for intelligence. "
    },
    {
      "id": "qwen/qwen3-vl-235b-a22b-thinking",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 VL 235B A22B Thinking",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000012",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video. The Thinking model is optimized for multimodal reasoning in STEM and math. The series emphasizes robust perception (recognition of diverse real-world and synthetic categories), spatial understanding (2D/3D grounding), and long-form visual comprehension, with competitive results on public multimodal benchmarks for both perception and reasoning.\n\nBeyond analysis, Qwen3-VL supports agentic interaction and tool use: it can follow complex instructions over multi-image, multi-turn dialogues; align text to video timelines for precise temporal queries; and operate GUI elements for automation tasks. The models also enable visual coding workflows, turning sketches or mockups into code and assisting with UI debugging, while maintaining strong text-only performance comparable to the flagship Qwen3 language models. This makes Qwen3-VL suitable for production scenarios spanning document AI, multilingual OCR, software/UI assistance, spatial/embodied tasks, and research on vision-language agents."
    },
    {
      "id": "qwen/qwen3-vl-235b-a22b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 VL 235B A22B Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000022",
        "completion": "0.00000088",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video. The Instruct model targets general vision-language use (VQA, document parsing, chart/table extraction, multilingual OCR). The series emphasizes robust perception (recognition of diverse real-world and synthetic categories), spatial understanding (2D/3D grounding), and long-form visual comprehension, with competitive results on public multimodal benchmarks for both perception and reasoning.\n\nBeyond analysis, Qwen3-VL supports agentic interaction and tool use: it can follow complex instructions over multi-image, multi-turn dialogues; align text to video timelines for precise temporal queries; and operate GUI elements for automation tasks. The models also enable visual coding workflows—turning sketches or mockups into code and assisting with UI debugging—while maintaining strong text-only performance comparable to the flagship Qwen3 language models. This makes Qwen3-VL suitable for production scenarios spanning document AI, multilingual OCR, software/UI assistance, spatial/embodied tasks, and research on vision-language agents."
    },
    {
      "id": "qwen/qwen3-max",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Max",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000012",
        "completion": "0.000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000024"
      },
      "contextLength": 256000,
      "description": "Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the January 2025 version. It delivers higher accuracy in math, coding, logic, and science tasks, follows complex instructions in Chinese and English more reliably, reduces hallucinations, and produces higher-quality responses for open-ended Q&A, writing, and conversation. The model supports over 100 languages with stronger translation and commonsense reasoning, and is optimized for retrieval-augmented generation (RAG) and tool calling, though it does not include a dedicated “thinking” mode."
    },
    {
      "id": "qwen/qwen3-coder-plus",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Coder Plus",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000001",
        "completion": "0.000005",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000001"
      },
      "contextLength": 128000,
      "description": "Qwen3 Coder Plus is Alibaba's proprietary version of the Open Source Qwen3 Coder 480B A35B. It is a powerful coding agent model specializing in autonomous programming via tool calling and environment interaction, combining coding proficiency with versatile general-purpose abilities."
    },
    {
      "id": "openai/gpt-5-codex",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5 Codex",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000125",
        "completion": "0.00001",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000125"
      },
      "contextLength": 400000,
      "description": "GPT-5-Codex is a specialized version of GPT-5 optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks. The model supports building projects from scratch, feature development, debugging, large-scale refactoring, and code review. Compared to GPT-5, Codex is more steerable, adheres closely to developer instructions, and produces cleaner, higher-quality code outputs. Reasoning effort can be adjusted with the `reasoning.effort` parameter. Read the [docs here](https://openrouter.ai/docs/use-cases/reasoning-tokens#reasoning-effort-level)\n\nCodex integrates into developer environments including the CLI, IDE extensions, GitHub, and cloud tasks. It adapts reasoning effort dynamically—providing fast responses for small tasks while sustaining extended multi-hour runs for large projects. The model is trained to perform structured code reviews, catching critical flaws by reasoning over dependencies and validating behavior against tests. It also supports multimodal inputs such as images or screenshots for UI development and integrates tool use for search, dependency installation, and environment setup. Codex is intended specifically for agentic coding applications."
    },
    {
      "id": "qwen/qwen3-coder-flash",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Coder Flash",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000015",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000008"
      },
      "contextLength": 128000,
      "description": "Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus. It is a powerful coding agent model specializing in autonomous programming via tool calling and environment interaction, combining coding proficiency with versatile general-purpose abilities."
    },
    {
      "id": "qwen/qwen3-next-80b-a3b-thinking",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Next 80B A3B Thinking",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000015",
        "completion": "0.0000012",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default. It’s designed for hard multi-step problems; math proofs, code synthesis/debugging, logic, and agentic planning, and reports strong results across knowledge, reasoning, coding, alignment, and multilingual evaluations. Compared with prior Qwen3 variants, it emphasizes stability under long chains of thought and efficient scaling during inference, and it is tuned to follow complex instructions while reducing repetitive or off-task behavior.\n\nThe model is suitable for agent frameworks and tool use (function calling), retrieval-heavy workflows, and standardized benchmarking where step-by-step solutions are required. It supports long, detailed completions and leverages throughput-oriented techniques (e.g., multi-token prediction) for faster generation. Note that it operates in thinking-only mode."
    },
    {
      "id": "qwen/qwen3-next-80b-a3b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Next 80B A3B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000008",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks across reasoning, code generation, knowledge QA, and multilingual use, while remaining robust on alignment and formatting. Compared with prior Qwen3 instruct variants, it focuses on higher throughput and stability on ultra-long inputs and multi-turn dialogues, making it well-suited for RAG, tool use, and agentic workflows that require consistent final answers rather than visible chain-of-thought.\n\nThe model employs scaling-efficient training and decoding to improve parameter efficiency and inference speed, and has been validated on a broad set of public benchmarks where it reaches or approaches larger Qwen3 systems in several categories while outperforming earlier mid-sized baselines. It is best used as a general assistant, code helper, and long-context task solver in production settings where deterministic, instruction-following outputs are preferred."
    },
    {
      "id": "qwen/qwen-plus-2025-07-28",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen Plus 0728",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.0000012",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 1000000,
      "description": "Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination."
    },
    {
      "id": "qwen/qwen-plus-2025-07-28:thinking",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen Plus 0728 (thinking)",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 1000000,
      "description": "Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination."
    },
    {
      "id": "qwen/qwen3-30b-a3b-thinking-2507",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 30B A3B Thinking 2507",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000009",
        "completion": "0.0000003",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for “thinking mode,” where internal reasoning traces are separated from final answers.\n\nCompared to earlier Qwen3-30B releases, this version improves performance across logical reasoning, mathematics, science, coding, and multilingual benchmarks. It also demonstrates stronger instruction following, tool use, and alignment with human preferences. With higher reasoning efficiency and extended output budgets, it is best suited for advanced research, competitive problem solving, and agentic applications requiring structured long-context reasoning."
    },
    {
      "id": "google/gemini-2.5-flash-image-preview",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Flash Image Preview (Nano Banana)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000025",
        "request": "0",
        "image": "0.001238",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Gemini 2.5 Flash Image Preview, a.k.a. \"Nano Banana,\" is a state of the art image generation model with contextual understanding. It is capable of image generation, edits, and multi-turn conversations."
    },
    {
      "id": "openai/gpt-4o-audio-preview",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o Audio",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000025",
        "completion": "0.00001",
        "request": "0",
        "image": "0",
        "audio": "0.00004",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "The gpt-4o-audio-preview model adds support for audio inputs as prompts. This enhancement allows the model to detect nuances within audio recordings and add depth to generated user experiences. Audio outputs are currently not supported. Audio tokens are priced at $40 per million input audio tokens."
    },
    {
      "id": "mistralai/mistral-medium-3.1",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Medium 3.1",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost compared to traditional large models, making it suitable for scalable deployments across professional and industrial use cases.\n\nThe model excels in domains such as coding, STEM reasoning, and enterprise adaptation. It supports hybrid, on-prem, and in-VPC deployments and is optimized for integration into custom workflows. Mistral Medium 3.1 offers competitive accuracy relative to larger models like Claude Sonnet 3.5/3.7, Llama 4 Maverick, and Command R+, while maintaining broad compatibility across cloud environments."
    },
    {
      "id": "openai/gpt-5-chat",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5 Chat",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000125",
        "completion": "0.00001",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000125"
      },
      "contextLength": 128000,
      "description": "GPT-5 Chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications."
    },
    {
      "id": "openai/gpt-5",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000125",
        "completion": "0.00001",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000125"
      },
      "contextLength": 400000,
      "description": "GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. It supports test-time routing features and advanced prompt understanding, including user-specified intent like \"think hard about this.\" Improvements include reductions in hallucination, sycophancy, and better performance in coding, writing, and health-related tasks."
    },
    {
      "id": "openai/gpt-5-mini",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5 Mini",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000025",
        "completion": "0.000002",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000025"
      },
      "contextLength": 400000,
      "description": "GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost. GPT-5 Mini is the successor to OpenAI's o4-mini model."
    },
    {
      "id": "openai/gpt-5-nano",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-5 Nano",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000005"
      },
      "contextLength": 400000,
      "description": "GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments. While limited in reasoning depth compared to its larger counterparts, it retains key instruction-following and safety features. It is the successor to GPT-4.1-nano and offers a lightweight option for cost-sensitive or real-time applications."
    },
    {
      "id": "openai/gpt-oss-120b",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: gpt-oss-120b",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000004",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation."
    },
    {
      "id": "openai/gpt-oss-120b:exacto",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: gpt-oss-120b (exacto)",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.00000024",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation."
    },
    {
      "id": "openai/gpt-oss-20b:free",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: gpt-oss-20b (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference and deployability on consumer or single-GPU hardware. The model is trained in OpenAI’s Harmony response format and supports reasoning level configuration, fine-tuning, and agentic capabilities including function calling, tool use, and structured outputs."
    },
    {
      "id": "openai/gpt-oss-20b",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: gpt-oss-20b",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000003",
        "completion": "0.00000014",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference and deployability on consumer or single-GPU hardware. The model is trained in OpenAI’s Harmony response format and supports reasoning level configuration, fine-tuning, and agentic capabilities including function calling, tool use, and structured outputs."
    },
    {
      "id": "anthropic/claude-opus-4.1",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude Opus 4.1",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000015",
        "completion": "0.000075",
        "request": "0",
        "image": "0.024",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000015",
        "input_cache_write": "0.00001875"
      },
      "contextLength": 200000,
      "description": "Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks. It achieves 74.5% on SWE-bench Verified and shows notable gains in multi-file code refactoring, debugging precision, and detail-oriented reasoning. The model supports extended thinking up to 64K tokens and is optimized for tasks involving research, data analysis, and tool-assisted reasoning."
    },
    {
      "id": "mistralai/codestral-2508",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Codestral 2508",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000009",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 256000,
      "description": "Mistral's cutting-edge language model for coding released end of July 2025. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.\n\n[Blog Post](https://mistral.ai/news/codestral-25-08)"
    },
    {
      "id": "qwen/qwen3-coder-30b-a3b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Coder 30B A3B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000006",
        "completion": "0.00000025",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the Qwen3 architecture, it supports a native context length of 256K tokens (extendable to 1M with Yarn) and performs strongly in tasks involving function calls, browser use, and structured code completion.\n\nThis model is optimized for instruction-following without “thinking mode”, and integrates well with OpenAI-compatible tool-use formats. "
    },
    {
      "id": "qwen/qwen3-30b-a3b-instruct-2507",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 30B A3B Instruct 2507",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000008",
        "completion": "0.00000033",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and agentic tool use. Post-trained on instruction data, it demonstrates competitive performance across reasoning (AIME, ZebraLogic), coding (MultiPL-E, LiveCodeBench), and alignment (IFEval, WritingBench) benchmarks. It outperforms its non-instruct variant on subjective and open-ended tasks while retaining strong factual and coding performance."
    },
    {
      "id": "qwen/qwen3-235b-a22b-thinking-2507",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 235B A22B Thinking 2507",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000011",
        "completion": "0.0000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144 tokens of context. This \"thinking-only\" variant enhances structured logical reasoning, mathematics, science, and long-form generation, showing strong benchmark performance across AIME, SuperGPQA, LiveCodeBench, and MMLU-Redux. It enforces a special reasoning mode (</think>) and is designed for high-token outputs (up to 81,920 tokens) in challenging domains.\n\nThe model is instruction-tuned and excels at step-by-step reasoning, tool use, agentic workflows, and multilingual tasks. This release represents the most capable open-source variant in the Qwen3-235B series, surpassing many closed models in structured reasoning use cases."
    },
    {
      "id": "qwen/qwen3-coder:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Coder 480B A35B (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262000,
      "description": "Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over repositories. The model features 480 billion total parameters, with 35 billion active per forward pass (8 out of 160 experts).\n\nPricing for the Alibaba endpoints varies by context length. Once a request is greater than 128k input tokens, the higher pricing is used."
    },
    {
      "id": "qwen/qwen3-coder",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Coder 480B A35B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000022",
        "completion": "0.00000095",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over repositories. The model features 480 billion total parameters, with 35 billion active per forward pass (8 out of 160 experts).\n\nPricing for the Alibaba endpoints varies by context length. Once a request is greater than 128k input tokens, the higher pricing is used."
    },
    {
      "id": "qwen/qwen3-coder:exacto",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 Coder 480B A35B (exacto)",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000038",
        "completion": "0.00000153",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over repositories. The model features 480 billion total parameters, with 35 billion active per forward pass (8 out of 160 experts).\n\nPricing for the Alibaba endpoints varies by context length. Once a request is greater than 128k input tokens, the higher pricing is used."
    },
    {
      "id": "google/gemini-2.5-flash-lite",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Flash Lite",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000001",
        "input_cache_write": "0.0000001833"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance across common benchmarks compared to earlier Flash models. By default, \"thinking\" (i.e. multi-pass reasoning) is disabled to prioritize speed, but developers can enable it via the [Reasoning API parameter](https://openrouter.ai/docs/use-cases/reasoning-tokens) to selectively trade off cost for intelligence. "
    },
    {
      "id": "qwen/qwen3-235b-a22b-2507",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 235B A22B Instruct 2507",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000008",
        "completion": "0.00000055",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following, logical reasoning, math, code, and tool usage. The model supports a native 262K context length and does not implement \"thinking mode\" (<think> blocks).\n\nCompared to its base variant, this version delivers significant gains in knowledge coverage, long-context reasoning, coding benchmarks, and alignment with open-ended tasks. It is particularly strong on multilingual understanding, math reasoning (e.g., AIME, HMMT), and alignment evaluations like Arena-Hard and WritingBench."
    },
    {
      "id": "mistralai/devstral-medium",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Devstral Medium",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI. Positioned as a step up from Devstral Small, it achieves 61.6% on SWE-Bench Verified, placing it ahead of Gemini 2.5 Pro and GPT-4.1 in code-related tasks, at a fraction of the cost. It is designed for generalization across prompt styles and tool use in code agents and frameworks.\n\nDevstral Medium is available via API only (not open-weight), and supports enterprise deployment on private infrastructure, with optional fine-tuning capabilities."
    },
    {
      "id": "mistralai/devstral-small",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Devstral Small 1.1",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000007",
        "completion": "0.00000028",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "Devstral Small 1.1 is a 24B parameter open-weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI. Finetuned from Mistral Small 3.1 and released under the Apache 2.0 license, it features a 128k token context window and supports both Mistral-style function calling and XML output formats.\n\nDesigned for agentic coding workflows, Devstral Small 1.1 is optimized for tasks such as codebase exploration, multi-file edits, and integration into autonomous development agents like OpenHands and Cline. It achieves 53.6% on SWE-Bench Verified, surpassing all other open models on this benchmark, while remaining lightweight enough to run on a single 4090 GPU or Apple silicon machine. The model uses a Tekken tokenizer with a 131k vocabulary and is deployable via vLLM, Transformers, Ollama, LM Studio, and other OpenAI-compatible runtimes.\n"
    },
    {
      "id": "google/gemma-3n-e2b-it:free",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3n 2B (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8192,
      "description": "Gemma 3n E2B IT is a multimodal, instruction-tuned model developed by Google DeepMind, designed to operate efficiently at an effective parameter size of 2B while leveraging a 6B architecture. Based on the MatFormer architecture, it supports nested submodels and modular composition via the Mix-and-Match framework. Gemma 3n models are optimized for low-resource deployment, offering 32K context length and strong multilingual and reasoning performance across common benchmarks. This variant is trained on a diverse corpus including code, math, web, and multimodal data."
    },
    {
      "id": "mistralai/mistral-small-3.2-24b-instruct:free",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Small 3.2 24B (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling. Compared to the 3.1 release, version 3.2 significantly improves accuracy on WildBench and Arena Hard, reduces infinite generations, and delivers gains in tool use and structured output tasks.\n\nIt supports image and text inputs with structured outputs, function/tool calling, and strong performance across coding (HumanEval+, MBPP), STEM (MMLU, MATH, GPQA), and vision benchmarks (ChartQA, DocVQA)."
    },
    {
      "id": "mistralai/mistral-small-3.2-24b-instruct",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Small 3.2 24B",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000006",
        "completion": "0.00000018",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling. Compared to the 3.1 release, version 3.2 significantly improves accuracy on WildBench and Arena Hard, reduces infinite generations, and delivers gains in tool use and structured output tasks.\n\nIt supports image and text inputs with structured outputs, function/tool calling, and strong performance across coding (HumanEval+, MBPP), STEM (MMLU, MATH, GPQA), and vision benchmarks (ChartQA, DocVQA)."
    },
    {
      "id": "google/gemini-2.5-flash-lite-preview-06-17",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Flash Lite Preview 06-17",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "audio": "0.0000003",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000025",
        "input_cache_write": "0.0000001833"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance across common benchmarks compared to earlier Flash models. By default, \"thinking\" (i.e. multi-pass reasoning) is disabled to prioritize speed, but developers can enable it via the [Reasoning API parameter](https://openrouter.ai/docs/use-cases/reasoning-tokens) to selectively trade off cost for intelligence. "
    },
    {
      "id": "google/gemini-2.5-flash",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Flash",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000025",
        "request": "0",
        "image": "0.001238",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000003",
        "input_cache_write": "0.0000003833"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in \"thinking\" capabilities, enabling it to provide responses with greater accuracy and nuanced context handling. \n\nAdditionally, Gemini 2.5 Flash is configurable through the \"max tokens for reasoning\" parameter, as described in the documentation (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning)."
    },
    {
      "id": "google/gemini-2.5-pro",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Pro",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000125",
        "completion": "0.00001",
        "request": "0",
        "image": "0.00516",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000125",
        "input_cache_write": "0.000001625"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy and nuanced context handling. Gemini 2.5 Pro achieves top-tier performance on multiple benchmarks, including first-place positioning on the LMArena leaderboard, reflecting superior human-preference alignment and complex problem-solving abilities."
    },
    {
      "id": "openai/o3-pro",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o3 Pro",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00002",
        "completion": "0.00008",
        "request": "0",
        "image": "0.0153",
        "web_search": "0.01",
        "internal_reasoning": "0"
      },
      "contextLength": 200000,
      "description": "The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently better answers.\n\nNote that BYOK is required for this model. Set up here: https://openrouter.ai/settings/integrations"
    },
    {
      "id": "mistralai/magistral-small-2506",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Magistral Small 2506",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000005",
        "completion": "0.0000015",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40000,
      "description": "Magistral Small is a 24B parameter instruction-tuned model based on Mistral-Small-3.1 (2503), enhanced through supervised fine-tuning on traces from Magistral Medium and further refined via reinforcement learning. It is optimized for reasoning and supports a wide multilingual range, including over 20 languages."
    },
    {
      "id": "mistralai/magistral-medium-2506:thinking",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Magistral Medium 2506 (thinking)",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000005",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Magistral is Mistral's first reasoning model. It is ideal for general purpose use requiring longer thought processing and better accuracy than with non-reasoning LLMs. From legal research and financial forecasting to software development and creative storytelling — this model solves multi-step challenges where transparency and precision are critical."
    },
    {
      "id": "mistralai/magistral-medium-2506",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Magistral Medium 2506",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000005",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Magistral is Mistral's first reasoning model. It is ideal for general purpose use requiring longer thought processing and better accuracy than with non-reasoning LLMs. From legal research and financial forecasting to software development and creative storytelling — this model solves multi-step challenges where transparency and precision are critical."
    },
    {
      "id": "google/gemini-2.5-pro-preview",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Pro Preview 06-05",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000125",
        "completion": "0.00001",
        "request": "0",
        "image": "0.00516",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000031",
        "input_cache_write": "0.000001625"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy and nuanced context handling. Gemini 2.5 Pro achieves top-tier performance on multiple benchmarks, including first-place positioning on the LMArena leaderboard, reflecting superior human-preference alignment and complex problem-solving abilities.\n"
    },
    {
      "id": "anthropic/claude-opus-4",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude Opus 4",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000015",
        "completion": "0.000075",
        "request": "0",
        "image": "0.024",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000015",
        "input_cache_write": "0.00001875"
      },
      "contextLength": 200000,
      "description": "Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows. It sets new benchmarks in software engineering, achieving leading results on SWE-bench (72.5%) and Terminal-bench (43.2%). Opus 4 supports extended, agentic workflows, handling thousands of task steps continuously for hours without degradation. \n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-4)"
    },
    {
      "id": "anthropic/claude-sonnet-4",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude Sonnet 4",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000003",
        "completion": "0.000015",
        "request": "0",
        "image": "0.0048",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000003",
        "input_cache_write": "0.00000375"
      },
      "contextLength": 1000000,
      "description": "Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability. Achieving state-of-the-art performance on SWE-bench (72.7%), Sonnet 4 balances capability and computational efficiency, making it suitable for a broad range of applications from routine coding tasks to complex software development projects. Key enhancements include improved autonomous codebase navigation, reduced error rates in agent-driven workflows, and increased reliability in following intricate instructions. Sonnet 4 is optimized for practical everyday use, providing advanced reasoning capabilities while maintaining efficiency and responsiveness in diverse internal and external scenarios.\n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-4)"
    },
    {
      "id": "mistralai/devstral-small-2505",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Devstral Small 2505",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000006",
        "completion": "0.00000012",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "Devstral-Small-2505 is a 24B parameter agentic LLM fine-tuned from Mistral-Small-3.1, jointly developed by Mistral AI and All Hands AI for advanced software engineering tasks. It is optimized for codebase exploration, multi-file editing, and integration into coding agents, achieving state-of-the-art results on SWE-Bench Verified (46.8%).\n\nDevstral supports a 128k context window and uses a custom Tekken tokenizer. It is text-only, with the vision encoder removed, and is suitable for local deployment on high-end consumer hardware (e.g., RTX 4090, 32GB RAM Macs). Devstral is best used in agentic workflows via the OpenHands scaffold and is compatible with inference frameworks like vLLM, Transformers, and Ollama. It is released under the Apache 2.0 license."
    },
    {
      "id": "google/gemma-3n-e4b-it:free",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3n 4B (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8192,
      "description": "Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets. It supports multimodal inputs—including text, visual data, and audio—enabling diverse tasks such as text generation, speech recognition, translation, and image analysis. Leveraging innovations like Per-Layer Embedding (PLE) caching and the MatFormer architecture, Gemma 3n dynamically manages memory usage and computational load by selectively activating model parameters, significantly reducing runtime resource requirements.\n\nThis model supports a wide linguistic range (trained in over 140 languages) and features a flexible 32K token context window. Gemma 3n can selectively load parameters, optimizing memory and computational efficiency based on the task or device capabilities, making it well-suited for privacy-focused, offline-capable applications and on-device AI solutions. [Read more in the blog post](https://developers.googleblog.com/en/introducing-gemma-3n/)"
    },
    {
      "id": "google/gemma-3n-e4b-it",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3n 4B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000002",
        "completion": "0.00000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets. It supports multimodal inputs—including text, visual data, and audio—enabling diverse tasks such as text generation, speech recognition, translation, and image analysis. Leveraging innovations like Per-Layer Embedding (PLE) caching and the MatFormer architecture, Gemma 3n dynamically manages memory usage and computational load by selectively activating model parameters, significantly reducing runtime resource requirements.\n\nThis model supports a wide linguistic range (trained in over 140 languages) and features a flexible 32K token context window. Gemma 3n can selectively load parameters, optimizing memory and computational efficiency based on the task or device capabilities, making it well-suited for privacy-focused, offline-capable applications and on-device AI solutions. [Read more in the blog post](https://developers.googleblog.com/en/introducing-gemma-3n/)"
    },
    {
      "id": "openai/codex-mini",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: Codex Mini",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000015",
        "completion": "0.000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000375"
      },
      "contextLength": 200000,
      "description": "codex-mini-latest is a fine-tuned version of o4-mini specifically for use in Codex CLI. For direct use in the API, we recommend starting with gpt-4.1."
    },
    {
      "id": "meta-llama/llama-3.3-8b-instruct:free",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.3 8B Instruct (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "A lightweight and ultra-fast variant of Llama 3.3 70B, for use when quick response times are needed most."
    },
    {
      "id": "mistralai/mistral-medium-3",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Medium 3",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost compared to traditional large models, making it suitable for scalable deployments across professional and industrial use cases.\n\nThe model excels in domains such as coding, STEM reasoning, and enterprise adaptation. It supports hybrid, on-prem, and in-VPC deployments and is optimized for integration into custom workflows. Mistral Medium 3 offers competitive accuracy relative to larger models like Claude Sonnet 3.5/3.7, Llama 4 Maverick, and Command R+, while maintaining broad compatibility across cloud environments."
    },
    {
      "id": "google/gemini-2.5-pro-preview-05-06",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.5 Pro Preview 05-06",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000125",
        "completion": "0.00001",
        "request": "0",
        "image": "0.00516",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000031",
        "input_cache_write": "0.000001625"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy and nuanced context handling. Gemini 2.5 Pro achieves top-tier performance on multiple benchmarks, including first-place positioning on the LMArena leaderboard, reflecting superior human-preference alignment and complex problem-solving abilities."
    },
    {
      "id": "qwen/qwen3-4b:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 4B (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3-4B is a 4 billion parameter dense language model from the Qwen3 series, designed to support both general-purpose and reasoning-intensive tasks. It introduces a dual-mode architecture—thinking and non-thinking—allowing dynamic switching between high-precision logical reasoning and efficient dialogue generation. This makes it well-suited for multi-turn chat, instruction following, and complex agent workflows."
    },
    {
      "id": "meta-llama/llama-guard-4-12b",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama Guard 4 12B",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000018",
        "completion": "0.00000018",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 163840,
      "description": "Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM inputs (prompt classification) and in LLM responses (response classification). It acts as an LLM—generating text in its output that indicates whether a given prompt or response is safe or unsafe, and if unsafe, it also lists the content categories violated.\n\nLlama Guard 4 was aligned to safeguard against the standardized MLCommons hazards taxonomy and designed to support multimodal Llama 4 capabilities. Specifically, it combines features from previous Llama Guard models, providing content moderation for English and multiple supported languages, along with enhanced capabilities to handle mixed text-and-image prompts, including multiple images. Additionally, Llama Guard 4 is integrated into the Llama Moderations API, extending robust safety classification to text and images."
    },
    {
      "id": "qwen/qwen3-30b-a3b:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 30B A3B (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks. Its unique ability to switch seamlessly between a thinking mode for complex reasoning and a non-thinking mode for efficient dialogue ensures versatile, high-quality performance.\n\nSignificantly outperforming prior models like QwQ and Qwen2.5, Qwen3 delivers superior mathematics, coding, commonsense reasoning, creative writing, and interactive dialogue capabilities. The Qwen3-30B-A3B variant includes 30.5 billion parameters (3.3 billion activated), 48 layers, 128 experts (8 activated per task), and supports up to 131K token contexts with YaRN, setting a new standard among open-source models."
    },
    {
      "id": "qwen/qwen3-30b-a3b",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 30B A3B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000006",
        "completion": "0.00000022",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks. Its unique ability to switch seamlessly between a thinking mode for complex reasoning and a non-thinking mode for efficient dialogue ensures versatile, high-quality performance.\n\nSignificantly outperforming prior models like QwQ and Qwen2.5, Qwen3 delivers superior mathematics, coding, commonsense reasoning, creative writing, and interactive dialogue capabilities. The Qwen3-30B-A3B variant includes 30.5 billion parameters (3.3 billion activated), 48 layers, 128 experts (8 activated per task), and supports up to 131K token contexts with YaRN, setting a new standard among open-source models."
    },
    {
      "id": "qwen/qwen3-8b",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 8B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000000035",
        "completion": "0.000000138",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue. It supports seamless switching between \"thinking\" mode for math, coding, and logical inference, and \"non-thinking\" mode for general conversation. The model is fine-tuned for instruction-following, agent integration, creative writing, and multilingual use across 100+ languages and dialects. It natively supports a 32K token context window and can extend to 131K tokens with YaRN scaling."
    },
    {
      "id": "qwen/qwen3-14b:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 14B (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It supports seamless switching between a \"thinking\" mode for tasks like math, programming, and logical inference, and a \"non-thinking\" mode for general-purpose conversation. The model is fine-tuned for instruction-following, agent tool use, creative writing, and multilingual tasks across 100+ languages and dialects. It natively handles 32K token contexts and can extend to 131K tokens using YaRN-based scaling."
    },
    {
      "id": "qwen/qwen3-14b",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 14B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.00000022",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It supports seamless switching between a \"thinking\" mode for tasks like math, programming, and logical inference, and a \"non-thinking\" mode for general-purpose conversation. The model is fine-tuned for instruction-following, agent tool use, creative writing, and multilingual tasks across 100+ languages and dialects. It natively handles 32K token contexts and can extend to 131K tokens using YaRN-based scaling."
    },
    {
      "id": "qwen/qwen3-32b",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 32B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.0000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a \"thinking\" mode for tasks like math, coding, and logical inference, and a \"non-thinking\" mode for faster, general-purpose conversation. The model demonstrates strong performance in instruction-following, agent tool use, creative writing, and multilingual tasks across 100+ languages and dialects. It natively handles 32K token contexts and can extend to 131K tokens using YaRN-based scaling. "
    },
    {
      "id": "qwen/qwen3-235b-a22b:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 235B A22B (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass. It supports seamless switching between a \"thinking\" mode for complex reasoning, math, and code tasks, and a \"non-thinking\" mode for general conversational efficiency. The model demonstrates strong reasoning ability, multilingual support (100+ languages and dialects), advanced instruction-following, and agent tool-calling capabilities. It natively handles a 32K token context window and extends up to 131K tokens using YaRN-based scaling."
    },
    {
      "id": "qwen/qwen3-235b-a22b",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen3 235B A22B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000018",
        "completion": "0.00000054",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 40960,
      "description": "Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass. It supports seamless switching between a \"thinking\" mode for complex reasoning, math, and code tasks, and a \"non-thinking\" mode for general conversational efficiency. The model demonstrates strong reasoning ability, multilingual support (100+ languages and dialects), advanced instruction-following, and agent tool-calling capabilities. It natively handles a 32K token context window and extends up to 131K tokens using YaRN-based scaling."
    },
    {
      "id": "openai/o4-mini-high",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o4 Mini High",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000011",
        "completion": "0.0000044",
        "request": "0",
        "image": "0.0008415",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000275"
      },
      "contextLength": 200000,
      "description": "OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high. \n\nOpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning and coding performance across benchmarks like AIME (99.5% with Python) and SWE-bench, outperforming its predecessor o3-mini and even approaching o3 in some domains.\n\nDespite its smaller size, o4-mini exhibits high accuracy in STEM tasks, visual problem solving (e.g., MathVista, MMMU), and code editing. It is especially well-suited for high-throughput scenarios where latency or cost is critical. Thanks to its efficient architecture and refined reinforcement learning training, o4-mini can chain tools, generate structured outputs, and solve multi-step tasks with minimal delay—often in under a minute."
    },
    {
      "id": "openai/o3",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o3",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000008",
        "request": "0",
        "image": "0.00153",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000005"
      },
      "contextLength": 200000,
      "description": "o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following. Use it to think through multi-step problems that involve analysis across text, code, and images. "
    },
    {
      "id": "openai/o4-mini",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o4 Mini",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000011",
        "completion": "0.0000044",
        "request": "0",
        "image": "0.0008415",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000275"
      },
      "contextLength": 200000,
      "description": "OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning and coding performance across benchmarks like AIME (99.5% with Python) and SWE-bench, outperforming its predecessor o3-mini and even approaching o3 in some domains.\n\nDespite its smaller size, o4-mini exhibits high accuracy in STEM tasks, visual problem solving (e.g., MathVista, MMMU), and code editing. It is especially well-suited for high-throughput scenarios where latency or cost is critical. Thanks to its efficient architecture and refined reinforcement learning training, o4-mini can chain tools, generate structured outputs, and solve multi-step tasks with minimal delay—often in under a minute."
    },
    {
      "id": "qwen/qwen2.5-coder-7b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen2.5 Coder 7B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000003",
        "completion": "0.00000009",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing. Based on the Qwen2.5 architecture, it incorporates enhancements like RoPE, SwiGLU, RMSNorm, and GQA attention with support for up to 128K tokens using YaRN-based extrapolation. It is trained on a large corpus of source code, synthetic data, and text-code grounding, providing robust performance across programming languages and agentic coding workflows.\n\nThis model is part of the Qwen2.5-Coder family and offers strong compatibility with tools like vLLM for efficient deployment. Released under the Apache 2.0 license."
    },
    {
      "id": "openai/gpt-4.1",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4.1",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000008",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000005"
      },
      "contextLength": 1047576,
      "description": "GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning. It supports a 1 million token context window and outperforms GPT-4o and GPT-4.5 across coding (54.6% SWE-bench Verified), instruction compliance (87.4% IFEval), and multimodal understanding benchmarks. It is tuned for precise code diffs, agent reliability, and high recall in large document contexts, making it ideal for agents, IDE tooling, and enterprise knowledge retrieval."
    },
    {
      "id": "openai/gpt-4.1-mini",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4.1 Mini",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.0000016",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000001"
      },
      "contextLength": 1047576,
      "description": "GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost. It retains a 1 million token context window and scores 45.1% on hard instruction evals, 35.8% on MultiChallenge, and 84.1% on IFEval. Mini also shows strong coding ability (e.g., 31.6% on Aider’s polyglot diff benchmark) and vision understanding, making it suitable for interactive applications with tight performance constraints."
    },
    {
      "id": "openai/gpt-4.1-nano",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4.1 Nano",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000025"
      },
      "contextLength": 1047576,
      "description": "For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million token context window, and scores 80.1% on MMLU, 50.3% on GPQA, and 9.8% on Aider polyglot coding – even higher than GPT‑4o mini. It’s ideal for tasks like classification or autocompletion."
    },
    {
      "id": "meta-llama/llama-4-maverick:free",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 4 Maverick (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "Llama 4 Maverick 17B Instruct (128E) is a 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 pass (400B total). It supports multilingual text and image input, and produces multilingual text and code output across 12 supported languages. Optimized for vision-language tasks, Maverick is instruction-tuned for assistant-like behavior, image reasoning, and general-purpose multimodal interaction.\n\nMaverick features early fusion for native multimodality and a 1 million token context window. It was trained on a curated mixture of public, licensed, and Meta-platform data, covering ~22 trillion tokens, with a knowledge cutoff in August 2024. Released on April 5, 2025 under the Llama 4 Community License, Maverick is suited for research and commercial applications requiring advanced multimodal understanding and high model throughput."
    },
    {
      "id": "meta-llama/llama-4-maverick",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 4 Maverick",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000015",
        "completion": "0.0000006",
        "request": "0",
        "image": "0.0006684",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 1048576,
      "description": "Llama 4 Maverick 17B Instruct (128E) is a 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 pass (400B total). It supports multilingual text and image input, and produces multilingual text and code output across 12 supported languages. Optimized for vision-language tasks, Maverick is instruction-tuned for assistant-like behavior, image reasoning, and general-purpose multimodal interaction.\n\nMaverick features early fusion for native multimodality and a 1 million token context window. It was trained on a curated mixture of public, licensed, and Meta-platform data, covering ~22 trillion tokens, with a knowledge cutoff in August 2024. Released on April 5, 2025 under the Llama 4 Community License, Maverick is suited for research and commercial applications requiring advanced multimodal understanding and high model throughput."
    },
    {
      "id": "meta-llama/llama-4-scout:free",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 4 Scout (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, Scout uses 16 experts per forward pass and features a context length of 10 million tokens, with a training corpus of ~40 trillion tokens.\n\nBuilt for high efficiency and local or commercial deployment, Llama 4 Scout incorporates early fusion for seamless modality integration. It is instruction-tuned for use in multilingual chat, captioning, and image understanding tasks. Released under the Llama 4 Community License, it was last trained on data up to August 2024 and launched publicly on April 5, 2025."
    },
    {
      "id": "meta-llama/llama-4-scout",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 4 Scout",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000008",
        "completion": "0.0000003",
        "request": "0",
        "image": "0.0003342",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 327680,
      "description": "Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, Scout uses 16 experts per forward pass and features a context length of 10 million tokens, with a training corpus of ~40 trillion tokens.\n\nBuilt for high efficiency and local or commercial deployment, Llama 4 Scout incorporates early fusion for seamless modality integration. It is instruction-tuned for use in multilingual chat, captioning, and image understanding tasks. Released under the Llama 4 Community License, it was last trained on data up to August 2024 and launched publicly on April 5, 2025."
    },
    {
      "id": "qwen/qwen2.5-vl-32b-instruct:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen2.5 VL 32B Instruct (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 16384,
      "description": "Qwen2.5-VL-32B is a multimodal vision-language model fine-tuned through reinforcement learning for enhanced mathematical reasoning, structured outputs, and visual problem-solving capabilities. It excels at visual analysis tasks, including object recognition, textual interpretation within images, and precise event localization in extended videos. Qwen2.5-VL-32B demonstrates state-of-the-art performance across multimodal benchmarks such as MMMU, MathVista, and VideoMME, while maintaining strong reasoning and clarity in text-based tasks like MMLU, mathematical problem-solving, and code generation."
    },
    {
      "id": "qwen/qwen2.5-vl-32b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen2.5 VL 32B Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.00000022",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 16384,
      "description": "Qwen2.5-VL-32B is a multimodal vision-language model fine-tuned through reinforcement learning for enhanced mathematical reasoning, structured outputs, and visual problem-solving capabilities. It excels at visual analysis tasks, including object recognition, textual interpretation within images, and precise event localization in extended videos. Qwen2.5-VL-32B demonstrates state-of-the-art performance across multimodal benchmarks such as MMMU, MathVista, and VideoMME, while maintaining strong reasoning and clarity in text-based tasks like MMLU, mathematical problem-solving, and code generation."
    },
    {
      "id": "openai/o1-pro",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o1-pro",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00015",
        "completion": "0.0006",
        "request": "0",
        "image": "0.21675",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 200000,
      "description": "The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o1-pro model uses more compute to think harder and provide consistently better answers."
    },
    {
      "id": "mistralai/mistral-small-3.1-24b-instruct:free",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Small 3.1 24B (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 96000,
      "description": "Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities. It provides state-of-the-art performance in text-based reasoning and vision tasks, including image analysis, programming, mathematical reasoning, and multilingual support across dozens of languages. Equipped with an extensive 128k token context window and optimized for efficient local inference, it supports use cases such as conversational agents, function calling, long-document comprehension, and privacy-sensitive deployments. The updated version is [Mistral Small 3.2](mistralai/mistral-small-3.2-24b-instruct)"
    },
    {
      "id": "mistralai/mistral-small-3.1-24b-instruct",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Small 3.1 24B",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.0000001",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities. It provides state-of-the-art performance in text-based reasoning and vision tasks, including image analysis, programming, mathematical reasoning, and multilingual support across dozens of languages. Equipped with an extensive 128k token context window and optimized for efficient local inference, it supports use cases such as conversational agents, function calling, long-document comprehension, and privacy-sensitive deployments. The updated version is [Mistral Small 3.2](mistralai/mistral-small-3.2-24b-instruct)"
    },
    {
      "id": "google/gemma-3-4b-it:free",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3 4B (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling."
    },
    {
      "id": "google/gemma-3-4b-it",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3 4B",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000001703012",
        "completion": "0.0000000681536",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 96000,
      "description": "Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling."
    },
    {
      "id": "google/gemma-3-12b-it:free",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3 12B (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 12B is the second largest in the family of Gemma 3 models after [Gemma 3 27B](google/gemma-3-27b-it)"
    },
    {
      "id": "google/gemma-3-12b-it",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3 12B",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000003",
        "completion": "0.0000001",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 12B is the second largest in the family of Gemma 3 models after [Gemma 3 27B](google/gemma-3-27b-it)"
    },
    {
      "id": "openai/gpt-4o-mini-search-preview",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o-mini Search Preview",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000015",
        "completion": "0.0000006",
        "request": "0.0275",
        "image": "0.000217",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "GPT-4o mini Search Preview is a specialized model for web search in Chat Completions. It is trained to understand and execute web search queries."
    },
    {
      "id": "openai/gpt-4o-search-preview",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o Search Preview",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000025",
        "completion": "0.00001",
        "request": "0.035",
        "image": "0.003613",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "GPT-4o Search Previewis a specialized model for web search in Chat Completions. It is trained to understand and execute web search queries."
    },
    {
      "id": "google/gemma-3-27b-it:free",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3 27B (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 27B is Google's latest open source model, successor to [Gemma 2](google/gemma-2-27b-it)"
    },
    {
      "id": "google/gemma-3-27b-it",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 3 27B",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000009",
        "completion": "0.00000016",
        "request": "0",
        "image": "0.0000256",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 27B is Google's latest open source model, successor to [Gemma 2](google/gemma-2-27b-it)"
    },
    {
      "id": "qwen/qwq-32b",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: QwQ 32B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000015",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "QwQ is the reasoning model of the Qwen series. Compared with conventional instruction-tuned models, QwQ, which is capable of thinking and reasoning, can achieve significantly enhanced performance in downstream tasks, especially hard problems. QwQ-32B is the medium-sized reasoning model, which is capable of achieving competitive performance against state-of-the-art reasoning models, e.g., DeepSeek-R1, o1-mini."
    },
    {
      "id": "google/gemini-2.0-flash-lite-001",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.0 Flash Lite",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000000075",
        "completion": "0.0000003",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 1048576,
      "description": "Gemini 2.0 Flash Lite offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro 1.5](/google/gemini-pro-1.5), all at extremely economical token prices."
    },
    {
      "id": "anthropic/claude-3.7-sonnet:thinking",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3.7 Sonnet (thinking)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000003",
        "completion": "0.000015",
        "request": "0",
        "image": "0.0048",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000003",
        "input_cache_write": "0.00000375"
      },
      "contextLength": 200000,
      "description": "Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities. It introduces a hybrid reasoning approach, allowing users to choose between rapid responses and extended, step-by-step processing for complex tasks. The model demonstrates notable improvements in coding, particularly in front-end development and full-stack updates, and excels in agentic workflows, where it can autonomously navigate multi-step processes. \n\nClaude 3.7 Sonnet maintains performance parity with its predecessor in standard mode while offering an extended reasoning mode for enhanced accuracy in math, coding, and instruction-following tasks.\n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-3-7-sonnet)"
    },
    {
      "id": "anthropic/claude-3.7-sonnet",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3.7 Sonnet",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000003",
        "completion": "0.000015",
        "request": "0",
        "image": "0.0048",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000003",
        "input_cache_write": "0.00000375"
      },
      "contextLength": 200000,
      "description": "Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities. It introduces a hybrid reasoning approach, allowing users to choose between rapid responses and extended, step-by-step processing for complex tasks. The model demonstrates notable improvements in coding, particularly in front-end development and full-stack updates, and excels in agentic workflows, where it can autonomously navigate multi-step processes. \n\nClaude 3.7 Sonnet maintains performance parity with its predecessor in standard mode while offering an extended reasoning mode for enhanced accuracy in math, coding, and instruction-following tasks.\n\nRead more at the [blog post here](https://www.anthropic.com/news/claude-3-7-sonnet)"
    },
    {
      "id": "mistralai/mistral-saba",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Saba",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000002",
        "completion": "0.0000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. Trained on curated regional datasets, it supports multiple Indian-origin languages—including Tamil and Malayalam—alongside Arabic. This makes it a versatile option for a range of regional and multilingual applications. Read more at the blog post [here](https://mistral.ai/en/news/mistral-saba)"
    },
    {
      "id": "meta-llama/llama-guard-3-8b",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Llama Guard 3 8B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000002",
        "completion": "0.00000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Llama Guard 3 is a Llama-3.1-8B pretrained model, fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM inputs (prompt classification) and in LLM responses (response classification). It acts as an LLM – it generates text in its output that indicates whether a given prompt or response is safe or unsafe, and if unsafe, it also lists the content categories violated.\n\nLlama Guard 3 was aligned to safeguard against the MLCommons standardized hazards taxonomy and designed to support Llama 3.1 capabilities. Specifically, it provides content moderation in 8 languages, and was optimized to support safety and security for search and code interpreter tool calls.\n"
    },
    {
      "id": "openai/o3-mini-high",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o3 Mini High",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000011",
        "completion": "0.0000044",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000055"
      },
      "contextLength": 200000,
      "description": "OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high. \n\no3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding. The model features three adjustable reasoning effort levels and supports key developer capabilities including function calling, structured outputs, and streaming, though it does not include vision processing capabilities.\n\nThe model demonstrates significant improvements over its predecessor, with expert testers preferring its responses 56% of the time and noting a 39% reduction in major errors on complex questions. With medium reasoning effort settings, o3-mini matches the performance of the larger o1 model on challenging reasoning evaluations like AIME and GPQA, while maintaining lower latency and cost."
    },
    {
      "id": "google/gemini-2.0-flash-001",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.0 Flash",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000004",
        "request": "0",
        "image": "0.0000258",
        "audio": "0.0000007",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000025",
        "input_cache_write": "0.0000001833"
      },
      "contextLength": 1048576,
      "description": "Gemini Flash 2.0 offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro 1.5](/google/gemini-pro-1.5). It introduces notable enhancements in multimodal understanding, coding capabilities, complex instruction following, and function calling. These advancements come together to deliver more seamless and robust agentic experiences."
    },
    {
      "id": "qwen/qwen-vl-plus",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen VL Plus",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000021",
        "completion": "0.00000063",
        "request": "0",
        "image": "0.0002688",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 7500,
      "description": "Qwen's Enhanced Large Visual Language Model. Significantly upgraded for detailed recognition capabilities and text recognition abilities, supporting ultra-high pixel resolutions up to millions of pixels and extreme aspect ratios for image input. It delivers significant performance across a broad range of visual tasks.\n"
    },
    {
      "id": "qwen/qwen-vl-max",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen VL Max",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000008",
        "completion": "0.0000032",
        "request": "0",
        "image": "0.001024",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Qwen VL Max is a visual understanding model with 7500 tokens context length. It excels in delivering optimal performance for a broader spectrum of complex tasks.\n"
    },
    {
      "id": "qwen/qwen-turbo",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen-Turbo",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.0000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000002"
      },
      "contextLength": 1000000,
      "description": "Qwen-Turbo, based on Qwen2.5, is a 1M context model that provides fast speed and low cost, suitable for simple tasks."
    },
    {
      "id": "qwen/qwen2.5-vl-72b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen2.5 VL 72B Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000008",
        "completion": "0.00000033",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images."
    },
    {
      "id": "qwen/qwen-plus",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen-Plus",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.0000012",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000016"
      },
      "contextLength": 131072,
      "description": "Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination."
    },
    {
      "id": "qwen/qwen-max",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen-Max ",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000016",
        "completion": "0.0000064",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000064"
      },
      "contextLength": 32768,
      "description": "Qwen-Max, based on Qwen2.5, provides the best inference performance among [Qwen models](/qwen), especially for complex multi-step tasks. It's a large-scale MoE model that has been pretrained on over 20 trillion tokens and further post-trained with curated Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) methodologies. The parameter count is unknown."
    },
    {
      "id": "openai/o3-mini",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o3 Mini",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000011",
        "completion": "0.0000044",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000055"
      },
      "contextLength": 200000,
      "description": "OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding.\n\nThis model supports the `reasoning_effort` parameter, which can be set to \"high\", \"medium\", or \"low\" to control the thinking time of the model. The default is \"medium\". OpenRouter also offers the model slug `openai/o3-mini-high` to default the parameter to \"high\".\n\nThe model features three adjustable reasoning effort levels and supports key developer capabilities including function calling, structured outputs, and streaming, though it does not include vision processing capabilities.\n\nThe model demonstrates significant improvements over its predecessor, with expert testers preferring its responses 56% of the time and noting a 39% reduction in major errors on complex questions. With medium reasoning effort settings, o3-mini matches the performance of the larger o1 model on challenging reasoning evaluations like AIME and GPQA, while maintaining lower latency and cost."
    },
    {
      "id": "mistralai/mistral-small-24b-instruct-2501:free",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Small 3 (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed for efficient local deployment.\n\nThe model achieves 81% accuracy on the MMLU benchmark and performs competitively with larger models like Llama 3.3 70B and Qwen 32B, while operating at three times the speed on equivalent hardware. [Read the blog post about the model here.](https://mistral.ai/news/mistral-small-3/)"
    },
    {
      "id": "mistralai/mistral-small-24b-instruct-2501",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Small 3",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000005",
        "completion": "0.00000008",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed for efficient local deployment.\n\nThe model achieves 81% accuracy on the MMLU benchmark and performs competitively with larger models like Llama 3.3 70B and Qwen 32B, while operating at three times the speed on equivalent hardware. [Read the blog post about the model here.](https://mistral.ai/news/mistral-small-3/)"
    },
    {
      "id": "mistralai/codestral-2501",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Codestral 2501",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000009",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 262144,
      "description": "[Mistral](/mistralai)'s cutting-edge language model for coding. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation. \n\nLearn more on their blog post: https://mistral.ai/news/codestral-2501/"
    },
    {
      "id": "openai/o1",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: o1",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000015",
        "completion": "0.00006",
        "request": "0",
        "image": "0.021675",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000075"
      },
      "contextLength": 200000,
      "description": "The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding. The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought. \n\nThe o1 models are optimized for math, science, programming, and other STEM-related tasks. They consistently exhibit PhD-level accuracy on benchmarks in physics, chemistry, and biology. Learn more in the [launch announcement](https://openai.com/o1).\n"
    },
    {
      "id": "google/gemini-2.0-flash-exp:free",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemini 2.0 Flash Experimental (free)",
      "multiModal": true,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 1048576,
      "description": "Gemini Flash 2.0 offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro 1.5](/google/gemini-pro-1.5). It introduces notable enhancements in multimodal understanding, coding capabilities, complex instruction following, and function calling. These advancements come together to deliver more seamless and robust agentic experiences."
    },
    {
      "id": "meta-llama/llama-3.3-70b-instruct:free",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.3 70B Instruct (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model is optimized for multilingual dialogue use cases and outperforms many of the available open source and closed chat models on common industry benchmarks.\n\nSupported languages: English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai.\n\n[Model Card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md)"
    },
    {
      "id": "meta-llama/llama-3.3-70b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.3 70B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000013",
        "completion": "0.00000038",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model is optimized for multilingual dialogue use cases and outperforms many of the available open source and closed chat models on common industry benchmarks.\n\nSupported languages: English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai.\n\n[Model Card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md)"
    },
    {
      "id": "openai/gpt-4o-2024-11-20",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o (2024-11-20)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000025",
        "completion": "0.00001",
        "request": "0",
        "image": "0.003613",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000125"
      },
      "contextLength": 128000,
      "description": "The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability. It’s also better at working with uploaded files, providing deeper insights & more thorough responses.\n\nGPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities."
    },
    {
      "id": "mistralai/mistral-large-2411",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral Large 2411",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Mistral Large 2 2411 is an update of [Mistral Large 2](/mistralai/mistral-large) released together with [Pixtral Large 2411](/mistralai/pixtral-large-2411)\n\nIt provides a significant upgrade on the previous [Mistral Large 24.07](/mistralai/mistral-large-2407), with notable improvements in long context understanding, a new system prompt, and more accurate function calling."
    },
    {
      "id": "mistralai/mistral-large-2407",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral Large 2407",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/).\n\nIt supports dozens of languages including French, German, Spanish, Italian, Portuguese, Arabic, Hindi, Russian, Chinese, Japanese, and Korean, along with 80+ coding languages including Python, Java, C, C++, JavaScript, and Bash. Its long context window allows precise information recall from large documents.\n"
    },
    {
      "id": "mistralai/pixtral-large-2411",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Pixtral Large 2411",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000006",
        "request": "0",
        "image": "0.002888",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411). The model is able to understand documents, charts and natural images.\n\nThe model is available under the Mistral Research License (MRL) for research and educational use, and the Mistral Commercial License for experimentation, testing, and production for commercial purposes.\n\n"
    },
    {
      "id": "qwen/qwen-2.5-coder-32b-instruct:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen2.5 Coder 32B Instruct (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). Qwen2.5-Coder brings the following improvements upon CodeQwen1.5:\n\n- Significantly improvements in **code generation**, **code reasoning** and **code fixing**. \n- A more comprehensive foundation for real-world applications such as **Code Agents**. Not only enhancing coding capabilities but also maintaining its strengths in mathematics and general competencies.\n\nTo read more about its evaluation results, check out [Qwen 2.5 Coder's blog](https://qwenlm.github.io/blog/qwen2.5-coder-family/)."
    },
    {
      "id": "qwen/qwen-2.5-coder-32b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen2.5 Coder 32B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000004",
        "completion": "0.00000016",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). Qwen2.5-Coder brings the following improvements upon CodeQwen1.5:\n\n- Significantly improvements in **code generation**, **code reasoning** and **code fixing**. \n- A more comprehensive foundation for real-world applications such as **Code Agents**. Not only enhancing coding capabilities but also maintaining its strengths in mathematics and general competencies.\n\nTo read more about its evaluation results, check out [Qwen 2.5 Coder's blog](https://qwenlm.github.io/blog/qwen2.5-coder-family/)."
    },
    {
      "id": "anthropic/claude-3.5-haiku",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3.5 Haiku",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000008",
        "completion": "0.000004",
        "request": "0",
        "image": "0",
        "web_search": "0.01",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000008",
        "input_cache_write": "0.000001"
      },
      "contextLength": 200000,
      "description": "Claude 3.5 Haiku features offers enhanced capabilities in speed, coding accuracy, and tool use. Engineered to excel in real-time applications, it delivers quick response times that are essential for dynamic tasks such as chat interactions and immediate coding suggestions.\n\nThis makes it highly suitable for environments that demand both speed and precision, such as software development, customer service bots, and data management systems.\n\nThis model is currently pointing to [Claude 3.5 Haiku (2024-10-22)](/anthropic/claude-3-5-haiku-20241022)."
    },
    {
      "id": "anthropic/claude-3.5-haiku-20241022",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3.5 Haiku (2024-10-22)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000008",
        "completion": "0.000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000008",
        "input_cache_write": "0.000001"
      },
      "contextLength": 200000,
      "description": "Claude 3.5 Haiku features enhancements across all skill sets including coding, tool use, and reasoning. As the fastest model in the Anthropic lineup, it offers rapid response times suitable for applications that require high interactivity and low latency, such as user-facing chatbots and on-the-fly code completions. It also excels in specialized tasks like data extraction and real-time content moderation, making it a versatile tool for a broad range of industries.\n\nIt does not support image inputs.\n\nSee the launch announcement and benchmark results [here](https://www.anthropic.com/news/3-5-models-and-computer-use)"
    },
    {
      "id": "anthropic/claude-3.5-sonnet",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3.5 Sonnet",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000003",
        "completion": "0.000015",
        "request": "0",
        "image": "0.0048",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000003",
        "input_cache_write": "0.00000375"
      },
      "contextLength": 200000,
      "description": "New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices. Sonnet is particularly good at:\n\n- Coding: Scores ~49% on SWE-Bench Verified, higher than the last best score, and without any fancy prompt scaffolding\n- Data science: Augments human data science expertise; navigates unstructured data while using multiple tools for insights\n- Visual processing: excelling at interpreting charts, graphs, and images, accurately transcribing text to derive insights beyond just the text alone\n- Agentic tasks: exceptional tool use, making it great at agentic tasks (i.e. complex, multi-step problem solving tasks that require engaging with other systems)\n\n#multimodal"
    },
    {
      "id": "mistralai/ministral-3b",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Ministral 3B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000004",
        "completion": "0.00000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Ministral 3B is a 3B parameter model optimized for on-device and edge computing. It excels in knowledge, commonsense reasoning, and function-calling, outperforming larger models like Mistral 7B on most benchmarks. Supporting up to 128k context length, it’s ideal for orchestrating agentic workflows and specialist tasks with efficient inference."
    },
    {
      "id": "mistralai/ministral-8b",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Ministral 8B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000001",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Ministral 8B is an 8B parameter model featuring a unique interleaved sliding-window attention pattern for faster, memory-efficient inference. Designed for edge use cases, it supports up to 128k context length and excels in knowledge and reasoning tasks. It outperforms peers in the sub-10B category, making it perfect for low-latency, privacy-first applications."
    },
    {
      "id": "qwen/qwen-2.5-7b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen2.5 7B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000004",
        "completion": "0.0000001",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5 7B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2:\n\n- Significantly more knowledge and has greatly improved capabilities in coding and mathematics, thanks to our specialized expert models in these domains.\n\n- Significant improvements in instruction following, generating long texts (over 8K tokens), understanding structured data (e.g, tables), and generating structured outputs especially JSON. More resilient to the diversity of system prompts, enhancing role-play implementation and condition-setting for chatbots.\n\n- Long-context Support up to 128K tokens and can generate up to 8K tokens.\n\n- Multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.\n\nUsage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://huggingface.co/Qwen/Qwen1.5-110B-Chat/blob/main/LICENSE)."
    },
    {
      "id": "meta-llama/llama-3.2-1b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.2 1B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000000005",
        "completion": "0.00000001",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis. Its smaller size allows it to operate efficiently in low-resource environments while maintaining strong task performance.\n\nSupporting eight core languages and fine-tunable for more, Llama 1.3B is ideal for businesses or developers seeking lightweight yet powerful AI solutions that can operate in diverse multilingual settings without the high computational demand of larger models.\n\nClick here for the [original model card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/MODEL_CARD.md).\n\nUsage of this model is subject to [Meta's Acceptable Use Policy](https://www.llama.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3.2-90b-vision-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.2 90B Vision Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000035",
        "completion": "0.0000004",
        "request": "0",
        "image": "0.0005058",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "The Llama 90B Vision model is a top-tier, 90-billion-parameter multimodal model designed for the most challenging visual reasoning and language tasks. It offers unparalleled accuracy in image captioning, visual question answering, and advanced image-text comprehension. Pre-trained on vast multimodal datasets and fine-tuned with human feedback, the Llama 90B Vision is engineered to handle the most demanding image-based AI tasks.\n\nThis model is perfect for industries requiring cutting-edge multimodal AI capabilities, particularly those dealing with complex, real-time visual and textual analysis.\n\nClick here for the [original model card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/MODEL_CARD_VISION.md).\n\nUsage of this model is subject to [Meta's Acceptable Use Policy](https://www.llama.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3.2-3b-instruct:free",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.2 3B Instruct (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it supports eight languages, including English, Spanish, and Hindi, and is adaptable for additional languages.\n\nTrained on 9 trillion tokens, the Llama 3.2 3B model excels in instruction-following, complex reasoning, and tool use. Its balanced performance makes it ideal for applications needing accuracy and efficiency in text generation across multilingual settings.\n\nClick here for the [original model card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/MODEL_CARD.md).\n\nUsage of this model is subject to [Meta's Acceptable Use Policy](https://www.llama.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3.2-3b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.2 3B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000002",
        "completion": "0.00000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 16384,
      "description": "Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it supports eight languages, including English, Spanish, and Hindi, and is adaptable for additional languages.\n\nTrained on 9 trillion tokens, the Llama 3.2 3B model excels in instruction-following, complex reasoning, and tool use. Its balanced performance makes it ideal for applications needing accuracy and efficiency in text generation across multilingual settings.\n\nClick here for the [original model card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/MODEL_CARD.md).\n\nUsage of this model is subject to [Meta's Acceptable Use Policy](https://www.llama.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3.2-11b-vision-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.2 11B Vision Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000000049",
        "completion": "0.000000049",
        "request": "0",
        "image": "0.00007948",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Llama 3.2 11B Vision is a multimodal model with 11 billion parameters, designed to handle tasks combining visual and textual data. It excels in tasks such as image captioning and visual question answering, bridging the gap between language generation and visual reasoning. Pre-trained on a massive dataset of image-text pairs, it performs well in complex, high-accuracy image analysis.\n\nIts ability to integrate visual understanding with language processing makes it an ideal solution for industries requiring comprehensive visual-linguistic AI applications, such as content creation, AI-driven customer service, and research.\n\nClick here for the [original model card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/MODEL_CARD_VISION.md).\n\nUsage of this model is subject to [Meta's Acceptable Use Policy](https://www.llama.com/llama3/use-policy/)."
    },
    {
      "id": "qwen/qwen-2.5-72b-instruct:free",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen2.5 72B Instruct (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2:\n\n- Significantly more knowledge and has greatly improved capabilities in coding and mathematics, thanks to our specialized expert models in these domains.\n\n- Significant improvements in instruction following, generating long texts (over 8K tokens), understanding structured data (e.g, tables), and generating structured outputs especially JSON. More resilient to the diversity of system prompts, enhancing role-play implementation and condition-setting for chatbots.\n\n- Long-context Support up to 128K tokens and can generate up to 8K tokens.\n\n- Multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.\n\nUsage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://huggingface.co/Qwen/Qwen1.5-110B-Chat/blob/main/LICENSE)."
    },
    {
      "id": "qwen/qwen-2.5-72b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen2.5 72B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000007",
        "completion": "0.00000026",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2:\n\n- Significantly more knowledge and has greatly improved capabilities in coding and mathematics, thanks to our specialized expert models in these domains.\n\n- Significant improvements in instruction following, generating long texts (over 8K tokens), understanding structured data (e.g, tables), and generating structured outputs especially JSON. More resilient to the diversity of system prompts, enhancing role-play implementation and condition-setting for chatbots.\n\n- Long-context Support up to 128K tokens and can generate up to 8K tokens.\n\n- Multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.\n\nUsage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://huggingface.co/Qwen/Qwen1.5-110B-Chat/blob/main/LICENSE)."
    },
    {
      "id": "mistralai/pixtral-12b",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Pixtral 12B",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0.0000001",
        "request": "0",
        "image": "0.0001445",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "The first multi-modal, text+image-to-text model from Mistral AI. Its weights were launched via torrent: https://x.com/mistralai/status/1833758285167722836."
    },
    {
      "id": "qwen/qwen-2.5-vl-7b-instruct",
      "provider": "qwen",
      "providerId": "openrouter",
      "name": "Qwen: Qwen2.5-VL 7B Instruct",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000002",
        "completion": "0.0000002",
        "request": "0",
        "image": "0.0001445",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Qwen2.5 VL 7B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\n- SoTA understanding of images of various resolution & ratio: Qwen2.5-VL achieves state-of-the-art performance on visual understanding benchmarks, including MathVista, DocVQA, RealWorldQA, MTVQA, etc.\n\n- Understanding videos of 20min+: Qwen2.5-VL can understand videos over 20 minutes for high-quality video-based question answering, dialog, content creation, etc.\n\n- Agent that can operate your mobiles, robots, etc.: with the abilities of complex reasoning and decision making, Qwen2.5-VL can be integrated with devices like mobile phones, robots, etc., for automatic operation based on visual environment and text instructions.\n\n- Multilingual Support: to serve global users, besides English and Chinese, Qwen2.5-VL now supports the understanding of texts in different languages inside images, including most European languages, Japanese, Korean, Arabic, Vietnamese, etc.\n\nFor more details, see this [blog post](https://qwenlm.github.io/blog/qwen2-vl/) and [GitHub repo](https://github.com/QwenLM/Qwen2-VL).\n\nUsage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://huggingface.co/Qwen/Qwen1.5-110B-Chat/blob/main/LICENSE)."
    },
    {
      "id": "openai/chatgpt-4o-latest",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: ChatGPT-4o",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000005",
        "completion": "0.000015",
        "request": "0",
        "image": "0.007225",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "OpenAI ChatGPT 4o is continually updated by OpenAI to point to the current version of GPT-4o used by ChatGPT. It therefore differs slightly from the API version of [GPT-4o](/models/openai/gpt-4o) in that it has additional RLHF. It is intended for research and evaluation.\n\nOpenAI notes that this model is not suited for production use-cases as it may be removed or redirected to another model in the future."
    },
    {
      "id": "openai/gpt-4o-2024-08-06",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o (2024-08-06)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000025",
        "completion": "0.00001",
        "request": "0",
        "image": "0.003613",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000125"
      },
      "contextLength": 128000,
      "description": "The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format. Read more [here](https://openai.com/index/introducing-structured-outputs-in-the-api/).\n\nGPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)"
    },
    {
      "id": "meta-llama/llama-3.1-405b",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.1 405B (base)",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000004",
        "completion": "0.000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This is the base 405B pre-trained version.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3.1-70b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.1 70B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000004",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 70B instruct-tuned version is optimized for high quality dialogue usecases.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3-1/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3.1-8b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.1 8B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000002",
        "completion": "0.00000003",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 16384,
      "description": "Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3-1/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3.1-405b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3.1 405B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000008",
        "completion": "0.0000008",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "The highly anticipated 400B class of Llama3 is here! Clocking in at 128k context with impressive eval scores, the Meta AI team continues to push the frontier of open-source LLMs.\n\nMeta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 405B instruct-tuned version is optimized for high quality dialogue usecases.\n\nIt has demonstrated strong performance compared to leading closed-source models including GPT-4o and Claude 3.5 Sonnet in evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3-1/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/)."
    },
    {
      "id": "mistralai/mistral-nemo:free",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Nemo (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.\n\nThe model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, and Hindi.\n\nIt supports function calling and is released under the Apache 2.0 license."
    },
    {
      "id": "mistralai/mistral-nemo",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral Nemo",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000002",
        "completion": "0.00000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 131072,
      "description": "A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.\n\nThe model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, and Hindi.\n\nIt supports function calling and is released under the Apache 2.0 license."
    },
    {
      "id": "openai/gpt-4o-mini-2024-07-18",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o-mini (2024-07-18)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000015",
        "completion": "0.0000006",
        "request": "0",
        "image": "0.007225",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000075"
      },
      "contextLength": 128000,
      "description": "GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.\n\nAs their most advanced small model, it is many multiples more affordable than other recent frontier models, and more than 60% cheaper than [GPT-3.5 Turbo](/models/openai/gpt-3.5-turbo). It maintains SOTA intelligence, while being significantly more cost-effective.\n\nGPT-4o mini achieves an 82% score on MMLU and presently ranks higher than GPT-4 on chat preferences [common leaderboards](https://arena.lmsys.org/).\n\nCheck out the [launch announcement](https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/) to learn more.\n\n#multimodal"
    },
    {
      "id": "openai/gpt-4o-mini",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o-mini",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000015",
        "completion": "0.0000006",
        "request": "0",
        "image": "0.000217",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.000000075"
      },
      "contextLength": 128000,
      "description": "GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.\n\nAs their most advanced small model, it is many multiples more affordable than other recent frontier models, and more than 60% cheaper than [GPT-3.5 Turbo](/models/openai/gpt-3.5-turbo). It maintains SOTA intelligence, while being significantly more cost-effective.\n\nGPT-4o mini achieves an 82% score on MMLU and presently ranks higher than GPT-4 on chat preferences [common leaderboards](https://arena.lmsys.org/).\n\nCheck out the [launch announcement](https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/) to learn more.\n\n#multimodal"
    },
    {
      "id": "google/gemma-2-27b-it",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 2 27B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000065",
        "completion": "0.00000065",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8192,
      "description": "Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini).\n\nGemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning.\n\nSee the [launch announcement](https://blog.google/technology/developers/google-gemma-2/) for more details. Usage of Gemma is subject to Google's [Gemma Terms of Use](https://ai.google.dev/gemma/terms)."
    },
    {
      "id": "google/gemma-2-9b-it",
      "provider": "google",
      "providerId": "openrouter",
      "name": "Google: Gemma 2 9B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000003",
        "completion": "0.00000009",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8192,
      "description": "Gemma 2 9B by Google is an advanced, open-source language model that sets a new standard for efficiency and performance in its size class.\n\nDesigned for a wide variety of tasks, it empowers developers and researchers to build innovative applications, while maintaining accessibility, safety, and cost-effectiveness.\n\nSee the [launch announcement](https://blog.google/technology/developers/google-gemma-2/) for more details. Usage of Gemma is subject to Google's [Gemma Terms of Use](https://ai.google.dev/gemma/terms)."
    },
    {
      "id": "anthropic/claude-3.5-sonnet-20240620",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3.5 Sonnet (2024-06-20)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000003",
        "completion": "0.000015",
        "request": "0",
        "image": "0.0048",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000003",
        "input_cache_write": "0.00000375"
      },
      "contextLength": 200000,
      "description": "Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices. Sonnet is particularly good at:\n\n- Coding: Autonomously writes, edits, and runs code with reasoning and troubleshooting\n- Data science: Augments human data science expertise; navigates unstructured data while using multiple tools for insights\n- Visual processing: excelling at interpreting charts, graphs, and images, accurately transcribing text to derive insights beyond just the text alone\n- Agentic tasks: exceptional tool use, making it great at agentic tasks (i.e. complex, multi-step problem solving tasks that require engaging with other systems)\n\nFor the latest version (2024-10-23), check out [Claude 3.5 Sonnet](/anthropic/claude-3.5-sonnet).\n\n#multimodal"
    },
    {
      "id": "mistralai/mistral-7b-instruct:free",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral 7B Instruct (free)",
      "multiModal": false,
      "pricing": {
        "prompt": "0",
        "completion": "0",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "A high-performing, industry-standard 7.3B parameter model, with optimizations for speed and context length.\n\n*Mistral 7B Instruct has multiple version variants, and this is intended to be the latest version.*"
    },
    {
      "id": "mistralai/mistral-7b-instruct",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral 7B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000000028",
        "completion": "0.000000054",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "A high-performing, industry-standard 7.3B parameter model, with optimizations for speed and context length.\n\n*Mistral 7B Instruct has multiple version variants, and this is intended to be the latest version.*"
    },
    {
      "id": "mistralai/mistral-7b-instruct-v0.3",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral 7B Instruct v0.3",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000000028",
        "completion": "0.000000054",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "A high-performing, industry-standard 7.3B parameter model, with optimizations for speed and context length.\n\nAn improved version of [Mistral 7B Instruct v0.2](/models/mistralai/mistral-7b-instruct-v0.2), with the following changes:\n\n- Extended vocabulary to 32768\n- Supports v3 Tokenizer\n- Supports function calling\n\nNOTE: Support for function calling depends on the provider."
    },
    {
      "id": "meta-llama/llama-guard-2-8b",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: LlamaGuard 2 8B",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000002",
        "completion": "0.0000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8192,
      "description": "This safeguard model has 8B parameters and is based on the Llama 3 family. Just like is predecessor, [LlamaGuard 1](https://huggingface.co/meta-llama/LlamaGuard-7b), it can do both prompt and response classification.\n\nLlamaGuard 2 acts as a normal LLM would, generating text that indicates whether the given input/output is safe/unsafe. If deemed unsafe, it will also share the content categories violated.\n\nFor best results, please use raw prompt input or the `/completions` endpoint, instead of the chat API.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/)."
    },
    {
      "id": "openai/gpt-4o-2024-05-13",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o (2024-05-13)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000005",
        "completion": "0.000015",
        "request": "0",
        "image": "0.007225",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)\n\n#multimodal"
    },
    {
      "id": "openai/gpt-4o",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o",
      "multiModal": true,
      "pricing": {
        "prompt": "0.0000025",
        "completion": "0.00001",
        "request": "0",
        "image": "0.003613",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000125"
      },
      "contextLength": 128000,
      "description": "GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)\n\n#multimodal"
    },
    {
      "id": "openai/gpt-4o:extended",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4o (extended)",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000006",
        "completion": "0.000018",
        "request": "0",
        "image": "0.007225",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)\n\n#multimodal"
    },
    {
      "id": "meta-llama/llama-3-8b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3 8B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000003",
        "completion": "0.00000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8192,
      "description": "Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors. This 8B instruct-tuned version was optimized for high quality dialogue usecases.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/)."
    },
    {
      "id": "meta-llama/llama-3-70b-instruct",
      "provider": "meta-llama",
      "providerId": "openrouter",
      "name": "Meta: Llama 3 70B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000003",
        "completion": "0.0000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8192,
      "description": "Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors. This 70B instruct-tuned version was optimized for high quality dialogue usecases.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/)."
    },
    {
      "id": "mistralai/mixtral-8x22b-instruct",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mixtral 8x22B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 65536,
      "description": "Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include:\n- strong math, coding, and reasoning\n- large context length (64k)\n- fluency in English, French, Italian, German, and Spanish\n\nSee benchmarks on the launch announcement [here](https://mistral.ai/news/mixtral-8x22b/).\n#moe"
    },
    {
      "id": "openai/gpt-4-turbo",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4 Turbo",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00001",
        "completion": "0.00003",
        "request": "0",
        "image": "0.01445",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling.\n\nTraining data: up to December 2023."
    },
    {
      "id": "anthropic/claude-3-haiku",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3 Haiku",
      "multiModal": true,
      "pricing": {
        "prompt": "0.00000025",
        "completion": "0.00000125",
        "request": "0",
        "image": "0.0004",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.00000003",
        "input_cache_write": "0.0000003"
      },
      "contextLength": 200000,
      "description": "Claude 3 Haiku is Anthropic's fastest and most compact model for\nnear-instant responsiveness. Quick and accurate targeted performance.\n\nSee the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-haiku)\n\n#multimodal"
    },
    {
      "id": "anthropic/claude-3-opus",
      "provider": "anthropic",
      "providerId": "openrouter",
      "name": "Anthropic: Claude 3 Opus",
      "multiModal": true,
      "pricing": {
        "prompt": "0.000015",
        "completion": "0.000075",
        "request": "0",
        "image": "0.024",
        "web_search": "0",
        "internal_reasoning": "0",
        "input_cache_read": "0.0000015",
        "input_cache_write": "0.00001875"
      },
      "contextLength": 200000,
      "description": "Claude 3 Opus is Anthropic's most powerful model for highly complex tasks. It boasts top-level performance, intelligence, fluency, and understanding.\n\nSee the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-family)\n\n#multimodal"
    },
    {
      "id": "mistralai/mistral-large",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral Large",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000002",
        "completion": "0.000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/).\n\nIt supports dozens of languages including French, German, Spanish, Italian, Portuguese, Arabic, Hindi, Russian, Chinese, Japanese, and Korean, along with 80+ coding languages including Python, Java, C, C++, JavaScript, and Bash. Its long context window allows precise information recall from large documents."
    },
    {
      "id": "openai/gpt-4-turbo-preview",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4 Turbo Preview",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00001",
        "completion": "0.00003",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "The preview GPT-4 model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more. Training data: up to Dec 2023.\n\n**Note:** heavily rate limited by OpenAI while in preview."
    },
    {
      "id": "mistralai/mistral-tiny",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral Tiny",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000025",
        "completion": "0.00000025",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Note: This model is being deprecated. Recommended replacement is the newer [Ministral 8B](/mistral/ministral-8b)\n\nThis model is currently powered by Mistral-7B-v0.2, and incorporates a \"better\" fine-tuning than [Mistral 7B](/models/mistralai/mistral-7b-instruct-v0.1), inspired by community work. It's best used for large batch processing tasks where cost is a significant factor but reasoning capabilities are not crucial."
    },
    {
      "id": "mistralai/mistral-small",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral Small",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000002",
        "completion": "0.0000006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "With 22 billion parameters, Mistral Small v24.09 offers a convenient mid-point between (Mistral NeMo 12B)[/mistralai/mistral-nemo] and (Mistral Large 2)[/mistralai/mistral-large], providing a cost-effective solution that can be deployed across various platforms and environments. It has better reasoning, exhibits more capabilities, can produce and reason about code, and is multiligual, supporting English, French, German, Italian, and Spanish."
    },
    {
      "id": "mistralai/mistral-7b-instruct-v0.2",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral 7B Instruct v0.2",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000002",
        "completion": "0.0000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "A high-performing, industry-standard 7.3B parameter model, with optimizations for speed and context length.\n\nAn improved version of [Mistral 7B Instruct](/modelsmistralai/mistral-7b-instruct-v0.1), with the following changes:\n\n- 32k context window (vs 8k context in v0.1)\n- Rope-theta = 1e6\n- No Sliding-Window Attention"
    },
    {
      "id": "mistralai/mixtral-8x7b-instruct",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mixtral 8x7B Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000054",
        "completion": "0.00000054",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 32768,
      "description": "Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use. Incorporates 8 experts (feed-forward networks) for a total of 47 billion parameters.\n\nInstruct model fine-tuned by Mistral. #moe"
    },
    {
      "id": "openai/gpt-4-1106-preview",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4 Turbo (older v1106)",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00001",
        "completion": "0.00003",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 128000,
      "description": "The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling.\n\nTraining data: up to April 2023."
    },
    {
      "id": "openai/gpt-3.5-turbo-instruct",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-3.5 Turbo Instruct",
      "multiModal": false,
      "pricing": {
        "prompt": "0.0000015",
        "completion": "0.000002",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 4095,
      "description": "This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations. Training data: up to Sep 2021."
    },
    {
      "id": "mistralai/mistral-7b-instruct-v0.1",
      "provider": "mistralai",
      "providerId": "openrouter",
      "name": "Mistral: Mistral 7B Instruct v0.1",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00000011",
        "completion": "0.00000019",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 2824,
      "description": "A 7.3B parameter model that outperforms Llama 2 13B on all benchmarks, with optimizations for speed and context length."
    },
    {
      "id": "openai/gpt-3.5-turbo-16k",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-3.5 Turbo 16k",
      "multiModal": false,
      "pricing": {
        "prompt": "0.000003",
        "completion": "0.000004",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 16385,
      "description": "This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost. Training data: up to Sep 2021."
    },
    {
      "id": "openai/gpt-4-0314",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4 (older v0314)",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00003",
        "completion": "0.00006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8191,
      "description": "GPT-4-0314 is the first version of GPT-4 released, with a context length of 8,192 tokens, and was supported until June 14. Training data: up to Sep 2021."
    },
    {
      "id": "openai/gpt-4",
      "provider": "openai",
      "providerId": "openrouter",
      "name": "OpenAI: GPT-4",
      "multiModal": false,
      "pricing": {
        "prompt": "0.00003",
        "completion": "0.00006",
        "request": "0",
        "image": "0",
        "web_search": "0",
        "internal_reasoning": "0"
      },
      "contextLength": 8191,
      "description": "OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and advanced reasoning capabilities. Training data: up to Sep 2021."
    }
  ]
}