{"object":"list","data":[{"premium_output_modalities":[],"id":"anthropic/claude-fable-5-1","object":"model","created":1789344000,"released":1788220800,"expiration_date":null,"deprecated":false,"owned_by":"anthropic","name":"Claude Fable 5.1","description":"Anthropic's most capable model for demanding reasoning and long-running agent tasks.","context_window":1000000,"max_tokens":128000,"knowledge":"2026-06","modalities":{"input":["text","image"],"output":["text","search"]},"supported_surfaces":["messages"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00001","min":0}],"cache_read":{"tiers":[{"cost":"0.00000025","min":0}]},"cache_write":{"tiers":[{"cost":"0.0000125","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.00002","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.00005","min":0}]}},"service_tiers":{"us":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000011","min":0}],"cache_read":{"tiers":[{"cost":"0.000000275","min":0}]},"cache_write":{"tiers":[{"cost":"0.00001375","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.000022","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.000055","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"messages":"provider_retained"}},{"premium_output_modalities":[],"id":"anthropic/claude-haiku-4-5","object":"model","created":1789344000,"released":1760486400,"expiration_date":null,"deprecated":false,"owned_by":"anthropic","name":"Claude Haiku 4.5","description":"Anthropic's fastest model for straightforward tasks.","context_window":200000,"max_tokens":64000,"knowledge":"2025-02","modalities":{"input":["text","image"],"output":["text","search"]},"supported_surfaces":["messages"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000001","min":0}],"cache_read":{"tiers":[{"cost":"0.0000001","min":0}]},"cache_write":{"tiers":[{"cost":"0.00000125","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.000002","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.000005","min":0}]}}},"data_retention":"provider_retained","data_retention_by_surface":{"messages":"provider_retained"}},{"premium_output_modalities":[],"id":"anthropic/claude-opus-5","object":"model","created":1789344000,"released":1784851200,"expiration_date":null,"deprecated":false,"owned_by":"anthropic","name":"Claude Opus 5","description":"Anthropic's model for complex coding and agent tasks.","context_window":1000000,"max_tokens":128000,"knowledge":"2026-05","modalities":{"input":["text","image"],"output":["text","search"]},"supported_surfaces":["messages"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000005","min":0}],"cache_read":{"tiers":[{"cost":"0.0000005","min":0}]},"cache_write":{"tiers":[{"cost":"0.00000625","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.00001","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.000025","min":0}]}},"service_tiers":{"fast":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00001","min":0}],"cache_read":{"tiers":[{"cost":"0.000001","min":0}]},"cache_write":{"tiers":[{"cost":"0.0000125","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.00002","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.00005","min":0}]}}},"fast_us":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000011","min":0}],"cache_read":{"tiers":[{"cost":"0.0000011","min":0}]},"cache_write":{"tiers":[{"cost":"0.00001375","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.000022","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.000055","min":0}]}}},"us":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000055","min":0}],"cache_read":{"tiers":[{"cost":"0.00000055","min":0}]},"cache_write":{"tiers":[{"cost":"0.000006875","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.000011","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.0000275","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"messages":"provider_retained"}},{"premium_output_modalities":[],"id":"anthropic/claude-opus-5-5","object":"model","created":1790035200,"released":1790035200,"expiration_date":null,"deprecated":false,"owned_by":"anthropic","name":"Claude Opus 5.5","description":"Anthropic's model for long-running agentic coding and knowledge work.","context_window":1000000,"max_tokens":128000,"knowledge":"2026-06","modalities":{"input":["text","image"],"output":["text","search"]},"supported_surfaces":["messages"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000004","min":0}],"cache_read":{"tiers":[{"cost":"0.0000002","min":0}]},"cache_write":{"tiers":[{"cost":"0.000005","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.000008","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.00002","min":0}]}},"service_tiers":{"fast":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000008","min":0}],"cache_read":{"tiers":[{"cost":"0.0000004","min":0}]},"cache_write":{"tiers":[{"cost":"0.00001","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.000016","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.00004","min":0}]}}},"fast_us":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000088","min":0}],"cache_read":{"tiers":[{"cost":"0.00000044","min":0}]},"cache_write":{"tiers":[{"cost":"0.000011","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.0000176","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.000044","min":0}]}}},"us":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000044","min":0}],"cache_read":{"tiers":[{"cost":"0.00000022","min":0}]},"cache_write":{"tiers":[{"cost":"0.0000055","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.0000088","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.000022","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"messages":"provider_retained"}},{"premium_output_modalities":[],"id":"anthropic/claude-sonnet-5","object":"model","created":1789344000,"released":1782777600,"expiration_date":null,"deprecated":false,"owned_by":"anthropic","name":"Claude Sonnet 5","description":"Anthropic's model balancing speed and intelligence.","context_window":1000000,"max_tokens":128000,"knowledge":"2026-01","modalities":{"input":["text","image"],"output":["text","search"]},"supported_surfaces":["messages"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000002","min":0}],"cache_read":{"tiers":[{"cost":"0.0000002","min":0}]},"cache_write":{"tiers":[{"cost":"0.0000025","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.000004","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.00001","min":0}]}},"service_tiers":{"us":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000022","min":0}],"cache_read":{"tiers":[{"cost":"0.00000022","min":0}]},"cache_write":{"tiers":[{"cost":"0.00000275","min":0}]},"cache_write_1h":{"tiers":[{"cost":"0.0000044","min":0}]}}},"output":{"search":{"tiers":[{"cost":"0.01","min":0}],"unit":"request"},"text":{"tiers":[{"cost":"0.000011","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"messages":"provider_retained"}},{"premium_output_modalities":[],"id":"google/gemini-2.5-flash","object":"model","created":1779667200,"released":1742428800,"expiration_date":1792454400,"deprecated":true,"owned_by":"google","name":"Gemini 2.5 Flash","description":"Fast and efficient multimodal model for high-volume tasks.","context_window":1000000,"max_tokens":65536,"knowledge":"2025-01","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.000001","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000003","min":0}],"cache_read":{"tiers":[{"cost":"0.00000003","min":0}]}},"video":{"tiers":[{"cost":"0.0000003","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.0000025","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.0000018","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000054","min":0}],"cache_read":{"tiers":[{"cost":"0.000000054","min":0}]}},"video":{"tiers":[{"cost":"0.00000054","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.0000045","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-2.5-flash-image","object":"model","created":1779667200,"released":1756166400,"expiration_date":1805068800,"deprecated":true,"owned_by":"google","name":"Gemini 2.5 Flash Image","description":"Image generation and editing model for text and image prompts.","context_window":32768,"max_tokens":65536,"knowledge":"2025-01","modalities":{"input":["text","image","video"],"output":["text","image"]},"supported_surfaces":["chat-completions","image-generation"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000003","min":0}]},"video":{"tiers":[{"cost":"0.0000003","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.00003","min":0}]},"text":{"tiers":[{"cost":"0.0000025","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","image-generation":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-2.5-flash-lite","object":"model","created":1779667200,"released":1750118400,"expiration_date":1792454400,"deprecated":true,"owned_by":"google","name":"Gemini 2.5 Flash Lite","description":"Lightweight version of Gemini Flash for cost-effective processing.","context_window":1048576,"max_tokens":65536,"knowledge":"2025-01","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.0000003","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000001","min":0}],"cache_read":{"tiers":[{"cost":"0.00000001","min":0}]}},"video":{"tiers":[{"cost":"0.0000001","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.0000004","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.00000054","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000018","min":0}],"cache_read":{"tiers":[{"cost":"0.000000018","min":0}]}},"video":{"tiers":[{"cost":"0.00000018","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00000072","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-2.5-flash-lite-preview-tts","object":"model","created":1783382400,"released":1765411200,"expiration_date":null,"deprecated":true,"owned_by":"google","name":"Gemini 2.5 Flash Lite Preview TTS","description":"Lightweight preview Gemini text-to-speech model for high-volume audio.","context_window":8192,"max_tokens":16384,"modalities":{"input":["text"],"output":["audio"]},"supported_surfaces":["speech"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.0000005","min":0}]}},"output":{"audio":{"tiers":[{"cost":"0.00001","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"speech":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-2.5-flash-tts","object":"model","created":1783382400,"released":1759190400,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 2.5 Flash TTS","description":"Fast, cost-efficient Gemini text-to-speech model for natural audio output.","context_window":32768,"max_tokens":16384,"knowledge":"2025-01","modalities":{"input":["text"],"output":["audio"]},"supported_surfaces":["speech"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.0000005","min":0}]}},"output":{"audio":{"tiers":[{"cost":"0.00001","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"speech":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-2.5-pro","object":"model","created":1779667200,"released":1742428800,"expiration_date":1792454400,"deprecated":true,"owned_by":"google","name":"Gemini 2.5 Pro","description":"Google's strongest Gemini 2.5 model for complex reasoning and multimodal tasks.","context_window":1048576,"max_tokens":65536,"knowledge":"2025-01","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.00000125","min":0,"max":200000},{"cost":"0.0000025","min":200000}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000125","min":0,"max":200000},{"cost":"0.0000025","min":200000}],"cache_read":{"tiers":[{"cost":"0.000000125","min":0,"max":200000},{"cost":"0.00000025","min":200000}]}},"video":{"tiers":[{"cost":"0.00000125","min":0,"max":200000},{"cost":"0.0000025","min":200000}]}},"output":{"text":{"tiers":[{"cost":"0.00001","min":0,"max":200000},{"cost":"0.000015","min":200000}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.00000225","min":0,"max":200000},{"cost":"0.0000045","min":200000}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000225","min":0,"max":200000},{"cost":"0.0000045","min":200000}],"cache_read":{"tiers":[{"cost":"0.000000225","min":0,"max":200000},{"cost":"0.00000045","min":200000}]}},"video":{"tiers":[{"cost":"0.00000225","min":0,"max":200000},{"cost":"0.0000045","min":200000}]}},"output":{"text":{"tiers":[{"cost":"0.000018","min":0,"max":200000},{"cost":"0.000027","min":200000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero"}},{"premium_output_modalities":["audio"],"id":"google/gemini-2.5-pro-tts","object":"model","created":1783382400,"released":1759190400,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 2.5 Pro TTS","description":"Highest-quality Gemini text-to-speech model for expressive audio.","context_window":32768,"max_tokens":16384,"knowledge":"2025-01","modalities":{"input":["text"],"output":["audio"]},"supported_surfaces":["speech"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.000001","min":0}]}},"output":{"audio":{"tiers":[{"cost":"0.00002","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"speech":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-3-flash-preview","object":"model","created":1779667200,"released":1765929600,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3 Flash Preview","description":"Fast chat model with a strong balance of speed, capability, and cost.","context_window":1000000,"max_tokens":65000,"knowledge":"2025-01","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.000001","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000005","min":0}],"cache_read":{"tiers":[{"cost":"0.00000005","min":0}]}},"video":{"tiers":[{"cost":"0.0000005","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.000003","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.0000018","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000009","min":0}],"cache_read":{"tiers":[{"cost":"0.00000009","min":0}]}},"video":{"tiers":[{"cost":"0.0000009","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.0000054","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"chat-completions":"zero","interactions":"provider_retained"}},{"premium_output_modalities":["image"],"id":"google/gemini-3-pro-image","object":"model","created":1782086400,"released":1779926400,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3 Pro Image","description":"Next-generation image generation model with higher quality output.","context_window":65536,"max_tokens":32768,"knowledge":"2025-01","modalities":{"input":["text","image","video"],"output":["text","image"]},"supported_surfaces":["chat-completions","image-generation"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000002","min":0}],"cache_read":{"tiers":[{"cost":"0.0000002","min":0}]}},"video":{"tiers":[{"cost":"0.000002","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.00012","min":0}]},"text":{"tiers":[{"cost":"0.000012","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","image-generation":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-3.1-flash-image","object":"model","created":1782086400,"released":1779926400,"expiration_date":null,"deprecated":true,"owned_by":"google","name":"Gemini 3.1 Flash Image","description":"Fast image generation and editing model with pro-level quality.","context_window":131072,"max_tokens":32768,"knowledge":"2025-01","modalities":{"input":["text","image","video"],"output":["text","image"]},"supported_surfaces":["chat-completions","image-generation"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000005","min":0}],"cache_read":{"tiers":[{"cost":"0.00000005","min":0}]}},"video":{"tiers":[{"cost":"0.0000005","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.00006","min":0}]},"text":{"tiers":[{"cost":"0.000003","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","image-generation":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-3.1-flash-lite","object":"model","created":1781654400,"released":1778112000,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3.1 Flash Lite","description":"Cost-efficient Gemini 3.1 model for high-volume chat and extraction tasks.","context_window":1000000,"max_tokens":65000,"knowledge":"2025-01","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.0000005","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000025","min":0}],"cache_read":{"tiers":[{"cost":"0.000000025","min":0}]}},"video":{"tiers":[{"cost":"0.00000025","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.0000015","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.0000009","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000045","min":0}],"cache_read":{"tiers":[{"cost":"0.000000045","min":0}]}},"video":{"tiers":[{"cost":"0.00000045","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.0000027","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"chat-completions":"zero","interactions":"provider_retained"}},{"premium_output_modalities":[],"id":"google/gemini-3.1-flash-lite-image","object":"model","created":1783296000,"released":1782777600,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3.1 Flash Lite Image","description":"Cost-efficient image generation and editing model.","context_window":65536,"max_tokens":4096,"modalities":{"input":["text","image","video"],"output":["text","image"]},"supported_surfaces":["image-generation"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000025","min":0}],"cache_read":{"tiers":[{"cost":"0.000000025","min":0}]}},"video":{"tiers":[{"cost":"0.00000025","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.00003","min":0}]},"text":{"tiers":[{"cost":"0.0000015","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"image-generation":"zero"}},{"premium_output_modalities":["audio"],"id":"google/gemini-3.1-flash-tts-preview","object":"model","created":1783382400,"released":1776211200,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3.1 Flash TTS Preview","description":"Preview of the latest-generation Gemini text-to-speech model.","context_window":8192,"max_tokens":16384,"knowledge":"2025-01","modalities":{"input":["text"],"output":["audio"]},"supported_surfaces":["speech"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.000001","min":0}]}},"output":{"audio":{"tiers":[{"cost":"0.00002","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"speech":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-3.1-pro-preview","object":"model","created":1779667200,"released":1771459200,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3.1 Pro Preview","description":"Latest preview of Google's next-generation reasoning model.","context_window":1000000,"max_tokens":64000,"knowledge":"2025-01","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.000002","min":0,"max":200000},{"cost":"0.000004","min":200000}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000002","min":0,"max":200000},{"cost":"0.000004","min":200000}],"cache_read":{"tiers":[{"cost":"0.0000002","min":0,"max":200000},{"cost":"0.0000004","min":200000}]}},"video":{"tiers":[{"cost":"0.000002","min":0,"max":200000},{"cost":"0.000004","min":200000}]}},"output":{"text":{"tiers":[{"cost":"0.000012","min":0,"max":200000},{"cost":"0.000018","min":200000}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.0000036","min":0,"max":200000},{"cost":"0.0000072","min":200000}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000036","min":0,"max":200000},{"cost":"0.0000072","min":200000}],"cache_read":{"tiers":[{"cost":"0.00000036","min":0,"max":200000},{"cost":"0.00000072","min":200000}]}},"video":{"tiers":[{"cost":"0.0000036","min":0,"max":200000},{"cost":"0.0000072","min":200000}]}},"output":{"text":{"tiers":[{"cost":"0.0000216","min":0,"max":200000},{"cost":"0.0000324","min":200000}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"chat-completions":"zero","interactions":"provider_retained"}},{"premium_output_modalities":[],"id":"google/gemini-3.5-flash","object":"model","created":1779667200,"released":1779148800,"expiration_date":null,"deprecated":true,"owned_by":"google","name":"Gemini 3.5 Flash","description":"High-efficiency Gemini 3.5 model for fast coding, reasoning, and agentic workflows.","context_window":1000000,"max_tokens":64000,"knowledge":"2025-01","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.0000015","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000015","min":0}],"cache_read":{"tiers":[{"cost":"0.00000015","min":0}]}},"video":{"tiers":[{"cost":"0.0000015","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.000009","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.0000027","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000027","min":0}],"cache_read":{"tiers":[{"cost":"0.00000027","min":0}]}},"video":{"tiers":[{"cost":"0.0000027","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.0000162","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero"}},{"premium_output_modalities":["text"],"id":"google/gemini-3.5-transcribe","object":"model","created":1787875200,"released":1787702400,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3.5 Transcribe","description":"Google's speech-to-text model with automatic language detection across 85+ languages and smart handling of filler words and self-corrections.","context_window":16000,"max_tokens":16000,"modalities":{"input":["text","audio"],"output":["text"]},"supported_surfaces":["transcription"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.000002","min":0}]},"text":{"tiers":[{"cost":"0.000002","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.000012","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"transcription":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-3.6-flash","object":"model","created":1784505600,"released":1784592000,"expiration_date":1795046400,"deprecated":true,"owned_by":"google","name":"Gemini 3.6 Flash","description":"Gemini 3.6 Flash model for fast coding, reasoning, and agentic workflows.","context_window":1000000,"max_tokens":64000,"knowledge":"2026-03","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.00000075","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000075","min":0}],"cache_read":{"tiers":[{"cost":"0.000000075","min":0}]}},"video":{"tiers":[{"cost":"0.00000075","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00000375","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.00000135","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000135","min":0}],"cache_read":{"tiers":[{"cost":"0.000000135","min":0}]}},"video":{"tiers":[{"cost":"0.00000135","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00000675","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"chat-completions":"zero","interactions":"provider_retained"}},{"premium_output_modalities":[],"id":"google/gemini-3.7-flash","object":"model","created":1786752000,"released":1786752000,"expiration_date":1801094400,"deprecated":true,"owned_by":"google","name":"Gemini 3.7 Flash","description":"Gemini 3.7 Flash model for fast coding, reasoning, and agentic workflows.","context_window":1000000,"max_tokens":64000,"knowledge":"2026-03","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.00000075","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000075","min":0}],"cache_read":{"tiers":[{"cost":"0.000000075","min":0}]}},"video":{"tiers":[{"cost":"0.00000075","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00000375","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.00000135","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000135","min":0}],"cache_read":{"tiers":[{"cost":"0.000000135","min":0}]}},"video":{"tiers":[{"cost":"0.00000135","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00000675","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"chat-completions":"zero","interactions":"provider_retained"}},{"premium_output_modalities":[],"id":"google/gemini-3.8-flash","object":"model","created":1788739200,"released":1788307200,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 3.8 Flash","description":"Latest-generation Gemini Flash model for fast coding, reasoning, and agentic workflows.","context_window":1048576,"max_tokens":65536,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.00000075","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000075","min":0}],"cache_read":{"tiers":[{"cost":"0.000000075","min":0}]}},"video":{"tiers":[{"cost":"0.00000075","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00000375","min":0}]}},"service_tiers":{"priority":{"input":{"audio":{"tiers":[{"cost":"0.00000135","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000135","min":0}],"cache_read":{"tiers":[{"cost":"0.000000135","min":0}]}},"video":{"tiers":[{"cost":"0.00000135","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00000675","min":0}]}}}}},"data_retention":"provider_retained","data_retention_by_surface":{"chat-completions":"zero","interactions":"provider_retained"}},{"premium_output_modalities":[],"id":"google/gemini-4-argon","object":"model","created":1791849600,"released":1791849600,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini 4 Argon","description":"Frontier Gemini model for long-horizon coding, reasoning, and enterprise knowledge work.","context_window":1048576,"max_tokens":1048576,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["chat-completions"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.000002","min":0}]},"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000002","min":0}],"cache_read":{"tiers":[{"cost":"0.0000001","min":0}]}},"video":{"tiers":[{"cost":"0.000002","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00001","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero"}},{"premium_output_modalities":[],"id":"google/gemini-embedding-001","object":"model","created":1779667200,"released":1747699200,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini Embedding 001","description":"Legacy text-only predecessor; only for existing tables that already store its vectors.","context_window":2048,"max_tokens":0,"knowledge":"2025-05","modalities":{"input":["text"],"output":["text"]},"supported_surfaces":["embeddings"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.00000015","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"embeddings":"zero"}},{"premium_output_modalities":["text"],"id":"google/gemini-embedding-2","object":"model","created":1781136000,"released":1776816000,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini Embedding 2","description":"Default embedding model for semantic search, recommendations, and RAG; multimodal (text, image, audio, and video inputs).","context_window":8192,"max_tokens":0,"knowledge":"2025-11","modalities":{"input":["text","image","audio","video"],"output":["text"]},"supported_surfaces":["embeddings"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.0000065","min":0}]},"image":{"tiers":[{"cost":"0.00000045","min":0}]},"text":{"tiers":[{"cost":"0.0000002","min":0}]},"video":{"tiers":[{"cost":"0.000012","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"embeddings":"zero"}},{"premium_output_modalities":["video"],"id":"google/gemini-omni-1.1-flash","object":"model","created":1787788800,"released":1787788800,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Gemini Omni 1.1 Flash","description":"Latest video generation model: fast and low-cost, 360p drafts up to 4K, 3-10 second clips, video editing and extension from inline inputs.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image","audio","video"],"output":["video"]},"supported_surfaces":["videos"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.0000015","min":0}]},"image":{"tiers":[{"cost":"0.0000015","min":0}]},"text":{"tiers":[{"cost":"0.0000015","min":0}]},"video":{"tiers":[{"cost":"0.0000015","min":0}]}},"video_duration":{"720p":{"cost_per_second":"0.0000175"}}},"data_retention":"provider_retained","data_retention_by_surface":{"videos":"provider_retained"}},{"premium_output_modalities":[],"id":"google/lyria-3-clip-preview","object":"model","created":1790640000,"released":1774396800,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Lyria 3 Clip Preview","description":"Generate 30-second music clips with vocals or instrumentals from text or images.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["audio"]},"supported_surfaces":["music"],"pricing":{"output":{"audio":{"tiers":[{"cost":"0.04","min":0}],"unit":"request"}}},"data_retention":"zero","data_retention_by_surface":{"music":"zero"}},{"premium_output_modalities":[],"id":"google/lyria-3-pro-preview","object":"model","created":1790640000,"released":1774396800,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Lyria 3 Pro Preview","description":"Generate full music tracks up to 184 seconds with vocals or instrumentals from text or images.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["audio"]},"supported_surfaces":["music"],"pricing":{"output":{"audio":{"tiers":[{"cost":"0.08","min":0}],"unit":"request"}}},"data_retention":"zero","data_retention_by_surface":{"music":"zero"}},{"premium_output_modalities":["video"],"id":"google/veo-3.1","object":"model","created":1785715200,"released":1760486400,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Veo 3.1","description":"Highest-quality Veo model for video generation.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["video"]},"supported_surfaces":["videos"],"pricing":{"video_duration":{"1080p":{"cost_per_second":"0.4"},"4k":{"cost_per_second":"0.6"},"720p":{"cost_per_second":"0.4"}}},"data_retention":"zero","data_retention_by_surface":{"videos":"zero"}},{"premium_output_modalities":[],"id":"google/veo-3.1-fast","object":"model","created":1785715200,"released":1760486400,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Veo 3.1 Fast","description":"Higher-quality video generation at moderate cost.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["video"]},"supported_surfaces":["videos"],"pricing":{"video_duration":{"1080p":{"cost_per_second":"0.12"},"4k":{"cost_per_second":"0.3"},"720p":{"cost_per_second":"0.1"}}},"data_retention":"zero","data_retention_by_surface":{"videos":"zero"}},{"premium_output_modalities":[],"id":"google/veo-3.1-lite","object":"model","created":1785715200,"released":1775088000,"expiration_date":null,"deprecated":false,"owned_by":"google","name":"Veo 3.1 Lite","description":"Cost-efficient video generation from a text prompt or an image.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["video"]},"supported_surfaces":["videos"],"pricing":{"video_duration":{"1080p":{"cost_per_second":"0.08"},"720p":{"cost_per_second":"0.05"}}},"data_retention":"zero","data_retention_by_surface":{"videos":"zero"}},{"premium_output_modalities":[],"id":"openai/chat-latest","object":"model","created":1787270400,"released":1754524800,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"ChatGPT Latest","description":"The ChatGPT model: points to the latest Instant model currently used in ChatGPT, tuned for natural conversational chat.","context_window":128000,"max_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000005","min":0}],"cache_read":{"tiers":[{"cost":"0.0000005","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.00003","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-4o-mini-transcribe","object":"model","created":1781568000,"released":1742428800,"expiration_date":1803600000,"deprecated":true,"owned_by":"openai","name":"GPT-4o mini Transcribe","description":"Fast, cost-efficient OpenAI speech-to-text model for dictation, voice input, and high-volume transcription.","context_window":16000,"max_tokens":2000,"modalities":{"input":["text","audio"],"output":["text"]},"supported_surfaces":["transcription"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.00000125","min":0}]},"text":{"tiers":[{"cost":"0.00000125","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.000005","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"transcription":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-4o-mini-tts","object":"model","created":1781568000,"released":1742428800,"expiration_date":1799193600,"deprecated":true,"owned_by":"openai","name":"GPT-4o mini TTS","description":"OpenAI text-to-speech model with instructions-steerable delivery (tone, accent, pacing) for natural-sounding audio.","context_window":2000,"max_tokens":0,"modalities":{"input":["text"],"output":["audio"]},"supported_surfaces":["speech"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.0000006","min":0}]}},"output":{"audio":{"tiers":[{"cost":"0.000012","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"speech":"zero"}},{"premium_output_modalities":["text"],"id":"openai/gpt-4o-transcribe","object":"model","created":1781654400,"released":1742428800,"expiration_date":1803600000,"deprecated":true,"owned_by":"openai","name":"GPT-4o Transcribe","description":"Higher-accuracy OpenAI speech-to-text model for noisy audio, accents, and specialist vocabulary.","context_window":16000,"max_tokens":2000,"modalities":{"input":["text","audio"],"output":["text"]},"supported_surfaces":["transcription"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.0000025","min":0}]},"text":{"tiers":[{"cost":"0.0000025","min":0}]}},"output":{"text":{"tiers":[{"cost":"0.00001","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"transcription":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5","object":"model","created":1779667200,"released":1754524800,"expiration_date":1796947200,"deprecated":true,"owned_by":"openai","name":"GPT-5","description":"Powerful all-rounder for reasoning, long-context, and multimodal tasks.","context_window":400000,"max_tokens":128000,"knowledge":"2024-09-30","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000125","min":0}],"cache_read":{"tiers":[{"cost":"0.000000125","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.00001","min":0}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000025","min":0}],"cache_read":{"tiers":[{"cost":"0.00000025","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.00002","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5-mini","object":"model","created":1779667200,"released":1754524800,"expiration_date":1796947200,"deprecated":true,"owned_by":"openai","name":"GPT-5 mini","description":"Smaller, faster GPT-5 model for efficient tasks.","context_window":400000,"max_tokens":128000,"knowledge":"2024-05-30","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000025","min":0}],"cache_read":{"tiers":[{"cost":"0.000000025","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.000002","min":0}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000045","min":0}],"cache_read":{"tiers":[{"cost":"0.000000045","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.0000036","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5-nano","object":"model","created":1779667200,"released":1754524800,"expiration_date":1796947200,"deprecated":true,"owned_by":"openai","name":"GPT-5 nano","description":"Most compact GPT-5 model for speed and cost efficiency.","context_window":400000,"max_tokens":128000,"knowledge":"2024-05-30","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000005","min":0}],"cache_read":{"tiers":[{"cost":"0.000000005","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.0000004","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.2","object":"model","created":1779667200,"released":1765411200,"expiration_date":null,"deprecated":true,"owned_by":"openai","name":"GPT-5.2","description":"OpenAI reasoning model for complex problem-solving tasks.","context_window":400000,"max_tokens":128000,"knowledge":"2024-10","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000175","min":0}],"cache_read":{"tiers":[{"cost":"0.000000175","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.000014","min":0}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000035","min":0}],"cache_read":{"tiers":[{"cost":"0.00000035","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.000028","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.4","object":"model","created":1779667200,"released":1772668800,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-5.4","description":"Advanced GPT-5.4 reasoning model for complex analysis and code generation.","context_window":1050000,"max_tokens":128000,"knowledge":"2025-08-31","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000025","min":0,"max":272000},{"cost":"0.000005","min":272000}],"cache_read":{"tiers":[{"cost":"0.00000025","min":0,"max":272000},{"cost":"0.0000005","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.000015","min":0,"max":272000},{"cost":"0.0000225","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000005","min":0}],"cache_read":{"tiers":[{"cost":"0.0000005","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.00003","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.4-mini","object":"model","created":1779667200,"released":1773705600,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-5.4 Mini","description":"Smaller, faster GPT-5.4 variant balancing reasoning quality and cost.","context_window":400000,"max_tokens":128000,"knowledge":"2025-08-31","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00000075","min":0}],"cache_read":{"tiers":[{"cost":"0.000000075","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.0000045","min":0}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000015","min":0}],"cache_read":{"tiers":[{"cost":"0.00000015","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.000009","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.4-nano","object":"model","created":1779667200,"released":1773705600,"expiration_date":1806537600,"deprecated":true,"owned_by":"openai","name":"GPT-5.4 Nano","description":"Fastest and lowest-cost GPT-5.4 variant for high-volume tasks.","context_window":400000,"max_tokens":128000,"knowledge":"2025-08-31","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000002","min":0}],"cache_read":{"tiers":[{"cost":"0.00000002","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.00000125","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.4-pro","object":"model","created":1779667200,"released":1772668800,"expiration_date":null,"deprecated":true,"owned_by":"openai","name":"GPT-5.4 Pro","description":"Premium GPT-5.4 variant with enhanced reasoning for complex tasks.","context_window":1050000,"max_tokens":128000,"knowledge":"2025-08-31","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00003","min":0,"max":272000},{"cost":"0.00006","min":272000}],"cache_read":{"tiers":[{"cost":"0.000003","min":0,"max":272000},{"cost":"0.000006","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00018","min":0,"max":272000},{"cost":"0.00027","min":272000}]}}},"data_retention":"zero","data_retention_by_surface":{"responses":"zero"}},{"premium_output_modalities":["text"],"id":"openai/gpt-5.5","object":"model","created":1779667200,"released":1776988800,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-5.5","description":"OpenAI's most capable GPT-5.5 model for demanding reasoning and coding tasks.","context_window":1000000,"max_tokens":128000,"knowledge":"2025-12-01","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000005","min":0,"max":272000},{"cost":"0.00001","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000005","min":0,"max":272000},{"cost":"0.000001","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00003","min":0,"max":272000},{"cost":"0.000045","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000125","min":0}],"cache_read":{"tiers":[{"cost":"0.00000125","min":0}]}}},"output":{"text":{"tiers":[{"cost":"0.000075","min":0}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.5-pro","object":"model","created":1783382400,"released":1776988800,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-5.5 Pro","description":"Premium GPT-5.5 variant with extended reasoning for the hardest problems.","context_window":1000000,"max_tokens":128000,"knowledge":"2025-12-01","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00003","min":0,"max":272000},{"cost":"0.00006","min":272000}],"cache_read":{"tiers":[{"cost":"0.00003","min":0,"max":272000},{"cost":"0.00006","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00018","min":0,"max":272000},{"cost":"0.00027","min":272000}]}}},"data_retention":"zero","data_retention_by_surface":{"responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.6-luna","object":"model","created":1784073600,"released":1783555200,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-5.6 Luna","description":"Fast, low-cost GPT-5.6 model for simple or high-volume tasks.","context_window":1050000,"max_tokens":128000,"knowledge":"2026-02-16","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000002","min":0,"max":272000},{"cost":"0.0000004","min":272000}],"cache_read":{"tiers":[{"cost":"0.00000002","min":0,"max":272000},{"cost":"0.00000004","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.0000012","min":0,"max":272000},{"cost":"0.0000018","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000004","min":0,"max":272000},{"cost":"0.0000008","min":272000}],"cache_read":{"tiers":[{"cost":"0.00000004","min":0,"max":272000},{"cost":"0.00000008","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.0000024","min":0,"max":272000},{"cost":"0.0000036","min":272000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":["text"],"id":"openai/gpt-5.6-sol","object":"model","created":1783987200,"released":1783555200,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-5.6 Sol","description":"OpenAI's flagship GPT-5.6 model for the hardest reasoning, coding, and agentic tasks.","context_window":1050000,"max_tokens":128000,"knowledge":"2026-02-16","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000004","min":0,"max":272000},{"cost":"0.000008","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000004","min":0,"max":272000},{"cost":"0.0000008","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00002","min":0,"max":272000},{"cost":"0.00003","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000008","min":0,"max":272000},{"cost":"0.000016","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000008","min":0,"max":272000},{"cost":"0.0000016","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00004","min":0,"max":272000},{"cost":"0.00006","min":272000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-5.6-terra","object":"model","created":1784073600,"released":1783555200,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for everyday work at lower cost.","context_window":1050000,"max_tokens":128000,"knowledge":"2026-02-16","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000002","min":0,"max":272000},{"cost":"0.000004","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000002","min":0,"max":272000},{"cost":"0.0000004","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.000012","min":0,"max":272000},{"cost":"0.000018","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000004","min":0,"max":272000},{"cost":"0.000008","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000004","min":0,"max":272000},{"cost":"0.0000008","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.000024","min":0,"max":272000},{"cost":"0.000036","min":272000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":["text"],"id":"openai/gpt-6-astra","object":"model","created":1788739200,"released":1788480000,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-6 Astra","description":"OpenAI's most capable model for complex reasoning, coding, research, and document creation.","context_window":1050000,"max_tokens":128000,"knowledge":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00001","min":0,"max":272000},{"cost":"0.00002","min":272000}],"cache_read":{"tiers":[{"cost":"0.000001","min":0,"max":272000},{"cost":"0.000002","min":272000}]},"cache_write":{"tiers":[{"cost":"0.0000125","min":0,"max":272000},{"cost":"0.000025","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00005","min":0,"max":272000},{"cost":"0.000075","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.00002","min":0,"max":272000},{"cost":"0.00004","min":272000}],"cache_read":{"tiers":[{"cost":"0.000002","min":0,"max":272000},{"cost":"0.000004","min":272000}]},"cache_write":{"tiers":[{"cost":"0.000025","min":0,"max":272000},{"cost":"0.00005","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.0001","min":0,"max":272000},{"cost":"0.00015","min":272000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-6-luna","object":"model","created":1790035200,"released":1790035200,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-6 Luna","description":"Low-cost GPT-6 model from OpenAI.","context_window":1050000,"max_tokens":128000,"knowledge":"2026-05-18","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000001","min":0,"max":272000},{"cost":"0.0000002","min":272000}],"cache_read":{"tiers":[{"cost":"0.00000001","min":0,"max":272000},{"cost":"0.00000002","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.0000005","min":0,"max":272000},{"cost":"0.00000075","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.0000002","min":0,"max":272000},{"cost":"0.0000004","min":272000}],"cache_read":{"tiers":[{"cost":"0.00000002","min":0,"max":272000},{"cost":"0.00000004","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.000001","min":0,"max":272000},{"cost":"0.0000015","min":272000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-6-sol","object":"model","created":1790035200,"released":1790035200,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-6 Sol","description":"Mid-tier GPT-6 model from OpenAI.","context_window":1050000,"max_tokens":128000,"knowledge":"2026-04-20","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000002","min":0,"max":272000},{"cost":"0.000004","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000002","min":0,"max":272000},{"cost":"0.0000004","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00001","min":0,"max":272000},{"cost":"0.000015","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000004","min":0,"max":272000},{"cost":"0.000008","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000004","min":0,"max":272000},{"cost":"0.0000008","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00002","min":0,"max":272000},{"cost":"0.00003","min":272000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-6.1-sol","object":"model","created":1791936000,"released":1790640000,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding and professional work. Use Responses for tool calling; Chat Completions supports requests without tools.","context_window":1050000,"max_tokens":128000,"knowledge":"2026-04-30","modalities":{"input":["text","image"],"output":["text"]},"supported_surfaces":["chat-completions","responses"],"pricing":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000002","min":0,"max":272000},{"cost":"0.000004","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000001","min":0,"max":272000},{"cost":"0.0000002","min":272000}]},"cache_write":{"tiers":[{"cost":"0.0000025","min":0,"max":272000},{"cost":"0.000005","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00001","min":0,"max":272000},{"cost":"0.000015","min":272000}]}},"service_tiers":{"priority":{"input":{"image":{"billed_as":"text"},"text":{"tiers":[{"cost":"0.000004","min":0,"max":272000},{"cost":"0.000008","min":272000}],"cache_read":{"tiers":[{"cost":"0.0000002","min":0,"max":272000},{"cost":"0.0000004","min":272000}]},"cache_write":{"tiers":[{"cost":"0.000005","min":0,"max":272000},{"cost":"0.00001","min":272000}]}}},"output":{"text":{"tiers":[{"cost":"0.00002","min":0,"max":272000},{"cost":"0.00003","min":272000}]}}}}},"data_retention":"zero","data_retention_by_surface":{"chat-completions":"zero","responses":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-image-1-mini","object":"model","created":1779753600,"released":1759708800,"expiration_date":1796083200,"deprecated":true,"owned_by":"openai","name":"GPT Image 1 Mini","description":"Cost-efficient OpenAI image generation model.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["image"]},"supported_surfaces":["image-generation","image-edit"],"pricing":{"input":{"image":{"tiers":[{"cost":"0.0000025","min":0}]},"text":{"tiers":[{"cost":"0.000002","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.000008","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"image-edit":"zero","image-generation":"zero"}},{"premium_output_modalities":["image"],"id":"openai/gpt-image-2","object":"model","created":1779753600,"released":1776729600,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT Image 2","description":"State-of-the-art OpenAI image generation model.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["image"]},"supported_surfaces":["image-generation","image-edit"],"pricing":{"input":{"image":{"tiers":[{"cost":"0.000008","min":0}]},"text":{"tiers":[{"cost":"0.000005","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.00003","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"image-edit":"zero","image-generation":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-image-2.5-flare","object":"model","created":1788912000,"released":1788825600,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT Image 2.5 Flare","description":"Latency-first OpenAI image generation and editing model.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["image"]},"supported_surfaces":["image-generation","image-edit"],"pricing":{"input":{"image":{"tiers":[{"cost":"0.000008","min":0}]},"text":{"tiers":[{"cost":"0.000005","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.00003","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"image-edit":"zero","image-generation":"zero"}},{"premium_output_modalities":["image"],"id":"openai/gpt-image-2.5-sunburst","object":"model","created":1788912000,"released":1788825600,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT Image 2.5 Sunburst","description":"Quality-first OpenAI image generation and editing model.","context_window":0,"max_tokens":0,"modalities":{"input":["text","image"],"output":["image"]},"supported_surfaces":["image-generation","image-edit"],"pricing":{"input":{"image":{"tiers":[{"cost":"0.000008","min":0}]},"text":{"tiers":[{"cost":"0.000005","min":0}]}},"output":{"image":{"tiers":[{"cost":"0.00003","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"image-edit":"zero","image-generation":"zero"}},{"premium_output_modalities":[],"id":"openai/gpt-live-1","object":"model","created":1789344000,"released":1788998400,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT-Live 1","description":"Low-latency voice conversations with interruptions and client delegation.","context_window":0,"max_tokens":0,"knowledge":"2025-07-31","modalities":{"input":["text","audio"],"output":["text","audio"]},"supported_surfaces":["live"],"pricing":{"output":{"audio":{"tiers":[{"cost":"0.000833333","min":0}],"unit":"second"}}},"data_retention":"zero","data_retention_by_surface":{"live":"zero"}},{"premium_output_modalities":["text"],"id":"openai/gpt-transcribe","object":"model","created":1789516800,"released":1785196800,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"GPT Transcribe","description":"Latest-generation OpenAI speech-to-text model with context, keyword, and language hints for higher accuracy on names and specialist vocabulary.","context_window":16000,"max_tokens":2000,"modalities":{"input":["text","audio"],"output":["text"]},"supported_surfaces":["transcription"],"pricing":{"input":{"audio":{"tiers":[{"cost":"0.000075","min":0}],"unit":"second"}}},"data_retention":"zero","data_retention_by_surface":{"transcription":"zero"}},{"premium_output_modalities":["text"],"id":"openai/text-embedding-3-large","object":"model","created":1779667200,"released":1706140800,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"text-embedding-3-large","description":"Higher-quality OpenAI embedding model when retrieval precision matters.","context_window":8191,"max_tokens":0,"modalities":{"input":["text"],"output":["text"]},"supported_surfaces":["embeddings"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.00000013","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"embeddings":"zero"}},{"premium_output_modalities":[],"id":"openai/text-embedding-3-small","object":"model","created":1779667200,"released":1706140800,"expiration_date":null,"deprecated":false,"owned_by":"openai","name":"text-embedding-3-small","description":"Cost-efficient OpenAI embedding model for high-volume retrieval workloads.","context_window":8191,"max_tokens":0,"modalities":{"input":["text"],"output":["text"]},"supported_surfaces":["embeddings"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.00000002","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"embeddings":"zero"}},{"premium_output_modalities":[],"id":"typesafe/jev-1.13","object":"model","created":1789689600,"released":1789689600,"expiration_date":null,"deprecated":false,"owned_by":"typesafe","name":"Jev 1.13","description":"Structured decisions from state and typed questions, returning probabilities for classification, scoring, and routing.","context_window":32000,"max_tokens":0,"modalities":{"input":["text"],"output":["text"]},"supported_surfaces":["systemone"],"pricing":{"input":{"text":{"tiers":[{"cost":"0.000000042","min":0}]}},"output":{"text":{"tiers":[{"cost":"0","min":0}]}}},"data_retention":"zero","data_retention_by_surface":{"systemone":"zero"}}]}