AI Providers
OpenRouter Models Access 100+ AI models from 15+ providers through OpenRouter
OpenRouter is an AI model aggregator that provides access to models from multiple providers through a single API. This gives you flexibility to use the best model for each task without managing multiple API keys.
API Key : Get your API key at openrouter.ai/keys
OpenRouter acts as a proxy to multiple AI providers:
Single API key for all models
Unified pricing
Automatic failover
Model comparison
Model format: provider/model-id
Model ID Name Description openai/gpt-5.1GPT-5.1 Latest GPT model openai/gpt-5.1-chatGPT-5.1 Chat Chat optimized openai/gpt-5.1-codexGPT-5.1 Codex Code focused openai/gpt-5GPT-5 Flagship model openai/gpt-5-miniGPT-5 Mini Compact version openai/gpt-4oGPT-4o Multimodal openai/gpt-4o-miniGPT-4o Mini Free tier openai/o3o3 Reasoning model openai/o3-minio3 Mini Compact reasoning openai/o4-minio4 Mini Latest reasoning openai/o1o1 Original reasoning
Model ID Name Description anthropic/claude-opus-4.5Claude Opus 4.5 Most capable anthropic/claude-opus-4.1Claude Opus 4.1 Advanced anthropic/claude-opus-4Claude Opus 4 High capability anthropic/claude-sonnet-4.5Claude Sonnet 4.5 Balanced anthropic/claude-3.5-sonnetClaude 3.5 Sonnet Efficient anthropic/claude-3-haikuClaude 3 Haiku Fast
Model ID Name Description google/gemini-3-pro-previewGemini 3 Pro Latest preview google/gemini-2.5-proGemini 2.5 Pro Most capable google/gemini-2.5-flashGemini 2.5 Flash Fast google/gemini-2.0-flash-001Gemini 2.0 Flash Efficient
Model ID Name Description x-ai/grok-4Grok 4 Latest Grok x-ai/grok-4-fastGrok 4 Fast Speed optimized x-ai/grok-4.1-fast:freeGrok 4.1 Fast Free tier x-ai/grok-3Grok 3 Previous gen
Model ID Name Description meta-llama/llama-4-scoutLlama 4 Scout Latest scout meta-llama/llama-4-maverickLlama 4 Maverick Latest maverick meta-llama/llama-3.3-70b-instructLlama 3.3 70B Large model meta-llama/llama-3.1-405b-instructLlama 3.1 405B Largest Llama meta-llama/llama-3.1-70b-instructLlama 3.1 70B Efficient
Model ID Name Description deepseek/deepseek-v3.2DeepSeek V3.2 Latest version deepseek/deepseek-v3.2-specialeDeepSeek V3.2 Speciale Enhanced deepseek/deepseek-r1DeepSeek R1 Reasoning deepseek/deepseek-chatDeepSeek Chat General chat
Model ID Name Description mistralai/mistral-large-2512Mistral Large 3 Latest large mistralai/mistral-large-2411Mistral Large 2 Previous gen
Model ID Name Description cohere/command-r-plus-08-2024Command R+ Most capable cohere/command-r-08-2024Command R Standard cohere/command-aCommand A Latest
Model ID Name Description qwen/qwen-2.5-72b-instructQwen 2.5 72B Large model qwen/qwen-2.5-coder-32b-instructQwen 2.5 Coder 32B Code focused qwen/qwen2.5-vl-72b-instructQwen 2.5 VL 72B Vision-language qwen/qwen3-coder-plusQwen3 Coder Plus Code focused
Model ID Name Description moonshotai/kimi-k2-thinkingKimi K2 Thinking Reasoning model moonshotai/kimi-k2-0905Kimi K2 Standard moonshotai/kimi-linear-48b-a3b-instructKimi Linear 48B Efficient
Model ID Name Description nvidia/llama-3.3-nemotron-super-49b-v1.5Nemotron Super 49B Enhanced Llama nvidia/nemotron-nano-12b-v2-vlNemotron Nano 12B VL Vision-language nvidia/nemotron-nano-9b-v2Nemotron Nano 9B Compact
Model ID Name Description allenai/olmo-3-32b-think:freeOlmo 3 32B Think Free reasoning allenai/olmo-3-7b-instructOlmo 3 7B Instruct Compact allenai/olmo-3-7b-thinkOlmo 3 7B Think Reasoning
Model ID Name Description perplexity/sonar-pro-searchSonar Pro Search Web search perplexity/sonar-proSonar Pro Standard perplexity/sonar-reasoning-proSonar Reasoning Pro Reasoning perplexity/sonarSonar Basic
Model ID Name Description google/gemini-3-pro-image-previewGemini 3 Pro Image Google's latest google/gemini-2.5-flash-imageGemini 2.5 Flash Image Fast openai/gpt-5-imageGPT-5 Image OpenAI latest openai/gpt-5-image-miniGPT-5 Image Mini Compact
Model ID Description openai/gpt-4o-miniGPT-4o Mini free tier openrouter/autoAuto-selects best free model x-ai/grok-4.1-fast:freeGrok free tier allenai/olmo-3-32b-think:freeOLMo free tier
Access all providers with one OpenRouter API key:
No need for multiple accounts
Unified billing
Simplified management
Switch between models easily:
// Configure different models in SheetMagic settings // No code changes required If one provider is down, OpenRouter can route to alternatives.
Compare pricing across providers
Use free tiers when appropriate
Optimize for cost vs. quality
// Model: openai/gpt-4o =AITEXT("Explain machine learning in simple terms") // Model: anthropic/claude-opus-4.5 =AITEXT("Write a detailed technical document about API design") // Model: openai/gpt-4o-mini (free) =AITEXT("Quick summary of: " & A1) Use Case Recommended Models General tasks openai/gpt-4o, anthropic/claude-sonnet-4.5 Complex reasoning openai/o3, deepseek/deepseek-r1 Code generation qwen/qwen3-coder-plus, openai/gpt-5.1-codex Cost-conscious openai/gpt-4o-mini, x-ai/grok-4.1-fast:free Research/search perplexity/sonar-pro-search
OpenRouter provides usage dashboards to track:
Token consumption
Costs by model
Request patterns
For simple tasks, use openrouter/auto to let OpenRouter select the optimal model.
OpenRouter charges based on:
Input tokens
Output tokens
Model selected
Pricing is often competitive with direct provider access, plus the convenience of unified billing.
OpenRouter sometimes offers credits for new users. Check their website for current promotions.
