{"data":[{"base_price":{"unit":"per_million_tokens","input":4,"output":7,"cache_read":0.4,"image_input":4,"image_output":7,"audio_input":4,"audio_output":7},"route_price_sources":{"z-official":"override"},"reference_price":{"unit":"per_million_tokens","input":1.4,"output":4.4,"cache_read":0.26},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://docs.z.ai/guides/overview/pricing","verified_at":"2026-09-18","model":"GLM-5.3"},"route_prices":{"z-official":{"unit":"per_million_tokens","input":4,"output":7,"cache_read":0.4,"image_input":4,"image_output":7,"audio_input":4,"audio_output":7}},"model_name":"GLM-5.3-Abliterated","reasoning":{"efforts":["low","high","max"],"default_effort":"max","wire_format":"chat"},"description":"GLM-5.3-Abliterated: a commercially fine-tuned variant of Z.ai's GLM-5.3 served from a private deployment, not the provider's official API. The tuning adds vertical business skills and removes the built-in refusal alignment (abliterated). Text only, with a 1M-token context, for harder reasoning and evaluation work; it always reasons, in three modes: low, high and max (the default). Priced above the Z.ai list price of the base model.","notice":"Because fine-tuning may affect safety alignment, the model may produce biased, offensive or locally unlawful content. Users bear full responsibility for compliance and legal consequences.","tags":"chat,tools,reasoning,1M context,128K output,fine-tuned,abliterated,private deployment","vendor_id":4,"quota_type":0,"owner_by":"","enable_groups":["z-official"],"supported_endpoint_types":["openai"],"auto_route":"z-official","context_window":1000000,"max_output":131072},{"base_price":{"unit":"per_million_tokens","input":10,"output":50,"cache_read":0.25,"cache_write":12.5,"cache_write_1h":20,"image_input":10,"image_output":50,"audio_input":10,"audio_output":50},"route_price_sources":{"claude-official":"override","claude-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":10,"output":50,"cache_read":0.25,"cache_write":12.5,"cache_write_1h":20},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://platform.claude.com/docs/en/about-claude/pricing","verified_at":"2026-09-24"},"route_prices":{"claude-official":{"unit":"per_million_tokens","input":9,"output":45,"cache_read":0.225,"cache_write":11.25,"cache_write_1h":18,"image_input":9,"image_output":45,"audio_input":9,"audio_output":45},"claude-shared":{"unit":"per_million_tokens","input":5,"output":25,"cache_read":0.125,"cache_write":6.25,"cache_write_1h":10,"image_input":5,"image_output":25,"audio_input":5,"audio_output":25}},"model_name":"claude-fable-5-1","reasoning":{"efforts":["low","medium","high","xhigh","max"],"default_effort":"high","wire_format":"chat"},"vendor_id":1,"quota_type":0,"owner_by":"","enable_groups":["claude-official","claude-shared"],"supported_endpoint_types":["anthropic","openai"],"auto_route":"claude-shared","context_window":1000000,"max_output":128000},{"base_price":{"unit":"per_million_tokens","input":4,"output":20,"cache_read":0.2,"cache_write":5,"cache_write_1h":8,"image_input":4,"image_output":20,"audio_input":4,"audio_output":20},"route_price_sources":{"claude-official":"override","claude-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":4,"output":20,"cache_read":0.2,"cache_write":5,"cache_write_1h":8},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://platform.claude.com/docs/en/about-claude/pricing","verified_at":"2026-09-24"},"route_prices":{"claude-official":{"unit":"per_million_tokens","input":3.6,"output":18,"cache_read":0.18,"cache_write":4.5,"cache_write_1h":7.2,"image_input":3.6,"image_output":18,"audio_input":3.6,"audio_output":18},"claude-shared":{"unit":"per_million_tokens","input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"cache_write_1h":4,"image_input":2,"image_output":10,"audio_input":2,"audio_output":10}},"model_name":"claude-opus-5-5","reasoning":{"efforts":["low","medium","high","xhigh","max"],"default_effort":"medium","wire_format":"chat"},"description":"Claude Opus 5.5: high-capability model for complex analysis and long-running agent tasks with a 1M-token context.","tags":"chat,vision,tools,reasoning,1M context,128K output","vendor_id":1,"quota_type":0,"owner_by":"","enable_groups":["claude-official","claude-shared"],"supported_endpoint_types":["anthropic","openai"],"auto_route":"claude-shared","context_window":1000000,"max_output":128000},{"base_price":{"unit":"per_million_tokens","input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"cache_write_1h":4,"image_input":2,"image_output":10,"audio_input":2,"audio_output":10},"route_price_sources":{"claude-official":"override","claude-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"cache_write_1h":4},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview","verified_at":"2026-10-02"},"route_prices":{"claude-official":{"unit":"per_million_tokens","input":1.8,"output":9,"cache_read":0.18,"cache_write":2.25,"cache_write_1h":3.6,"image_input":1.8,"image_output":9,"audio_input":1.8,"audio_output":9},"claude-shared":{"unit":"per_million_tokens","input":1,"output":5,"cache_read":0.1,"cache_write":1.25,"cache_write_1h":2,"image_input":1,"image_output":5,"audio_input":1,"audio_output":5}},"model_name":"claude-sonnet-5-5","reasoning":{"efforts":["low","medium","high","xhigh","max"],"default_effort":"high","wire_format":"chat"},"description":"Claude Sonnet 5.5: balanced speed, cost and intelligence for everyday workloads with a 1M-token context.","tags":"chat,vision,tools,1M context,128K output","vendor_id":1,"quota_type":0,"owner_by":"","enable_groups":["claude-official","claude-shared"],"supported_endpoint_types":["anthropic","openai"],"auto_route":"claude-shared","context_window":1000000,"max_output":128000},{"base_price":{"unit":"per_million_tokens","input":0.15,"output":0.6,"cache_read":0.003,"image_input":0.15,"image_output":0.6,"audio_input":0.15,"audio_output":0.6,"peak":{"schedule":"deepseek","input":0.3,"output":1.2,"cache_read":0.006,"image_input":0.3,"image_output":1.2,"audio_input":0.3,"audio_output":1.2}},"route_price_sources":{"deepseek-official":"override"},"reference_price":{"unit":"per_million_tokens","input":0.15,"output":0.6,"cache_read":0.003,"peak":{"schedule":"deepseek","input":0.3,"output":1.2,"cache_read":0.006}},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://api-docs.deepseek.com/quick_start/pricing","verified_at":"2026-09-22"},"route_prices":{"deepseek-official":{"unit":"per_million_tokens","input":0.12,"output":0.48,"cache_read":0.0024,"image_input":0.12,"image_output":0.48,"audio_input":0.12,"audio_output":0.48,"peak":{"schedule":"deepseek","input":0.24,"output":0.96,"cache_read":0.0048,"image_input":0.24,"image_output":0.96,"audio_input":0.24,"audio_output":0.96}}},"model_name":"deepseek-flash","reasoning":{"efforts":["none","low","high","max"],"default_effort":"high","wire_format":"deepseek"},"description":"DeepSeek-V4.1-Flash (deepseek-flash): DeepSeek's fast text and vision model with a 1M-token context and up to 384K output tokens, served on one official line at 80% of the DeepSeek list price. The provider's peak hours (01:00-04:00 and 06:00-10:00 UTC, Monday to Friday, outside Chinese public holidays) bill at twice the off-peak lanes.","tags":"chat,tools,vision,1M context,384K output,off-peak and peak","vendor_id":7,"quota_type":0,"owner_by":"","enable_groups":["deepseek-official"],"supported_endpoint_types":["openai"],"auto_route":"deepseek-official","context_window":1048576,"max_output":393216},{"base_price":{"unit":"per_million_tokens","input":0.66,"output":1.98,"cache_read":0.022,"image_input":0.66,"image_output":1.98,"audio_input":0.66,"audio_output":1.98,"peak":{"schedule":"deepseek","input":1.32,"output":3.96,"cache_read":0.044,"image_input":1.32,"image_output":3.96,"audio_input":1.32,"audio_output":3.96}},"route_price_sources":{"deepseek-official":"override"},"reference_price":{"unit":"per_million_tokens","input":0.66,"output":1.98,"cache_read":0.022,"peak":{"schedule":"deepseek","input":1.32,"output":3.96,"cache_read":0.044}},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://api-docs.deepseek.com/quick_start/pricing","verified_at":"2026-09-22"},"route_prices":{"deepseek-official":{"unit":"per_million_tokens","input":0.528,"output":1.584,"cache_read":0.0176,"image_input":0.528,"image_output":1.584,"audio_input":0.528,"audio_output":1.584,"peak":{"schedule":"deepseek","input":1.056,"output":3.168,"cache_read":0.0352,"image_input":1.056,"image_output":3.168,"audio_input":1.056,"audio_output":3.168}}},"model_name":"deepseek-v4-pro","reasoning":{"efforts":["none","low","high","max"],"default_effort":"high","wire_format":"deepseek"},"description":"DeepSeek-V4-Pro-0813 (deepseek-v4-pro): DeepSeek's strongest text model for reasoning, coding and agentic work, with a 1M-token context and up to 384K output tokens, served on one official line at 80% of the DeepSeek list price. The provider's peak hours (01:00-04:00 and 06:00-10:00 UTC, Monday to Friday, outside Chinese public holidays) bill at twice the off-peak lanes.","tags":"chat,tools,reasoning,1M context,384K output,off-peak and peak","vendor_id":7,"quota_type":0,"owner_by":"","enable_groups":["deepseek-official"],"supported_endpoint_types":["openai"],"auto_route":"deepseek-official","context_window":1048576,"max_output":393216},{"base_price":{"unit":"per_million_tokens","input":10,"output":50,"cache_read":1,"cache_write":12.5,"image_input":10,"image_output":50,"audio_input":10,"audio_output":50,"long_context":{"min_input_tokens":272001,"input":20,"output":75,"cache_read":2,"cache_write":25,"image_input":20,"image_output":75,"audio_input":20,"audio_output":75}},"route_price_sources":{"openai-official":"override","openai-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":10,"output":50,"cache_read":1,"cache_write":12.5,"long_context":{"min_input_tokens":272001,"input":20,"output":75,"cache_read":2,"cache_write":25}},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://developers.openai.com/api/docs/models/gpt-6-astra","verified_at":"2026-09-18"},"route_prices":{"openai-official":{"unit":"per_million_tokens","input":9,"output":45,"cache_read":0.9,"cache_write":11.25,"image_input":9,"image_output":45,"audio_input":9,"audio_output":45,"long_context":{"min_input_tokens":272001,"input":18,"output":67.5,"cache_read":1.8,"cache_write":22.5,"image_input":18,"image_output":67.5,"audio_input":18,"audio_output":67.5}},"openai-shared":{"unit":"per_million_tokens","input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"image_input":5,"image_output":25,"audio_input":5,"audio_output":25,"long_context":{"min_input_tokens":272001,"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"image_input":10,"image_output":37.5,"audio_input":10,"audio_output":37.5}}},"model_name":"gpt-6-astra","reasoning":{"efforts":["low","medium","high","xhigh","max"],"default_effort":"medium","wire_format":"responses"},"vendor_id":2,"quota_type":0,"owner_by":"","enable_groups":["openai-official","openai-shared"],"supported_endpoint_types":["openai","openai-response"],"auto_route":"openai-shared","context_window":1050000,"max_output":128000},{"base_price":{"unit":"per_million_tokens","input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"image_input":2,"image_output":10,"audio_input":2,"audio_output":10,"long_context":{"min_input_tokens":272001,"input":4,"output":15,"cache_read":0.2,"cache_write":5,"image_input":4,"image_output":15,"audio_input":4,"audio_output":15}},"route_price_sources":{"openai-official":"override","openai-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"long_context":{"min_input_tokens":272001,"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol","verified_at":"2026-09-30"},"route_prices":{"openai-official":{"unit":"per_million_tokens","input":1.8,"output":9,"cache_read":0.09,"cache_write":2.25,"image_input":1.8,"image_output":9,"audio_input":1.8,"audio_output":9,"long_context":{"min_input_tokens":272001,"input":3.6,"output":13.5,"cache_read":0.18,"cache_write":4.5,"image_input":3.6,"image_output":13.5,"audio_input":3.6,"audio_output":13.5}},"openai-shared":{"unit":"per_million_tokens","input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"image_input":1,"image_output":5,"audio_input":1,"audio_output":5,"long_context":{"min_input_tokens":272001,"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"image_input":2,"image_output":7.5,"audio_input":2,"audio_output":7.5}}},"model_name":"gpt-6.1-sol","reasoning":{"efforts":["low","medium","high","xhigh","max"],"default_effort":"medium","wire_format":"responses"},"description":"GPT-6.1 Sol: coding and agentic workflows with a 1.05M-token context and 128K output. Prompts with more than 272K input tokens are billed at the long-context rate (2x input and cache, 1.5x output) for the full request. Reasoning and tool calling are supported through Responses.","tags":"chat,vision,tools,reasoning,1.05M context,128K output,long-context tier","vendor_id":2,"quota_type":0,"owner_by":"","enable_groups":["openai-official","openai-shared"],"supported_endpoint_types":["openai","openai-response"],"auto_route":"openai-shared","context_window":1050000,"max_output":128000},{"base_price":{"unit":"per_image","sizes":{"1k":0.03,"2k":0.05}},"route_price_sources":{"image":"override"},"route_prices":{"image":{"unit":"per_image","sizes":{"1k":0.03,"2k":0.05}}},"model_name":"gpt-image-2.5-flare","reasoning":null,"description":"GPT Image 2.5 Flare: the faster GPT Image 2.5 model for image generation and editing, sold per image by output size (1K or 2K).","tags":"image,image-generation,image-edit,per image,1K,2K","vendor_id":2,"quota_type":0,"owner_by":"","enable_groups":["image"],"supported_endpoint_types":["image-generation","openai","image-edit"],"auto_route":"image"},{"base_price":{"unit":"per_image","sizes":{"1k":0.03,"2k":0.05}},"route_price_sources":{"image":"override"},"route_prices":{"image":{"unit":"per_image","sizes":{"1k":0.03,"2k":0.05}}},"model_name":"gpt-image-2.5-sunburst","reasoning":null,"description":"GPT Image 2.5 Sunburst: OpenAI's highest-fidelity image generation and editing model, sold per image by output size (1K or 2K).","tags":"image,image-generation,image-edit,per image,1K,2K","vendor_id":2,"quota_type":0,"owner_by":"","enable_groups":["image"],"supported_endpoint_types":["image-generation","openai","image-edit"],"auto_route":"image"},{"base_price":{"unit":"per_million_tokens","input":2,"output":6,"cache_read":0.5,"image_input":2,"image_output":6,"audio_input":2,"audio_output":6,"long_context":{"min_input_tokens":200000,"input":4,"output":12,"cache_read":1,"image_input":4,"image_output":12,"audio_input":4,"audio_output":12}},"route_price_sources":{"grok-official":"override","grok-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":2,"output":6,"cache_read":0.5,"long_context":{"min_input_tokens":200000,"input":4,"output":12,"cache_read":1}},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://docs.x.ai/developers/models/grok-4.7","verified_at":"2026-09-23"},"route_prices":{"grok-official":{"unit":"per_million_tokens","input":1.8,"output":5.4,"cache_read":0.45,"image_input":1.8,"image_output":5.4,"audio_input":1.8,"audio_output":5.4,"long_context":{"min_input_tokens":200000,"input":3.6,"output":10.8,"cache_read":0.9,"image_input":3.6,"image_output":10.8,"audio_input":3.6,"audio_output":10.8}},"grok-shared":{"unit":"per_million_tokens","input":1,"output":3,"cache_read":0.25,"image_input":1,"image_output":3,"audio_input":1,"audio_output":3,"long_context":{"min_input_tokens":200000,"input":2,"output":6,"cache_read":0.5,"image_input":2,"image_output":6,"audio_input":2,"audio_output":6}}},"model_name":"grok-4.7","reasoning":{"efforts":["low","medium","high","xhigh"],"default_effort":"high","wire_format":"chat"},"description":"Grok 4.7 by xAI: coding, agentic tasks and knowledge work with a 500K-token context. Prompts of 200K tokens or more are billed at the long-context rate (2x every lane) for the full request. Cached input is discounted and reasoning is enabled by default.","tags":"chat,vision,tools,reasoning,500K context,long-context tier","vendor_id":3,"quota_type":0,"owner_by":"","enable_groups":["grok-official","grok-shared"],"supported_endpoint_types":["openai","openai-response"],"auto_route":"grok-shared","context_window":500000},{"base_price":{"unit":"per_second","resolutions":{"1080p":0.2,"480p":0.05,"720p":0.1}},"route_price_sources":{"grok-official":"override"},"reference_price":{"unit":"per_second","unit_price":0.08},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://docs.x.ai/developers/pricing","verified_at":"2026-09-18"},"route_prices":{"grok-official":{"unit":"per_second","resolutions":{"1080p":0.2,"480p":0.05,"720p":0.1}}},"model_name":"grok-imagine-video-1.5","reasoning":null,"description":"Grok Imagine Video 1.5 by xAI: text-to-video and image-to-video generation, 1 to 15 seconds (default 8), billed per second by resolution (480p, 720p or 1080p).","tags":"video,video-generation,per second,480p,720p,1080p","vendor_id":3,"quota_type":0,"owner_by":"","enable_groups":["grok-official"],"supported_endpoint_types":["video-generation"],"auto_route":"grok-official"},{"base_price":{"unit":"per_million_tokens","input":0.042,"output":0,"cache_read":0.042,"image_input":0.042,"image_output":0,"audio_input":0.042,"audio_output":0},"route_price_sources":{"typesafe-official":"override"},"reference_price":{"unit":"per_million_tokens","input":0.042,"output":0,"cache_read":0.042},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://docs.typesafe.ai/models","verified_at":"2026-09-18"},"route_prices":{"typesafe-official":{"unit":"per_million_tokens","input":0.04494,"output":0,"cache_read":0.04494,"image_input":0.04494,"image_output":0,"audio_input":0.04494,"audio_output":0}},"model_name":"jev-1.13.0","reasoning":null,"description":"Jev 1.13.0 by TypeSafe: the first System One model. It answers typed questions (noul, choice, score) about a state with calibrated probabilities instead of generating text. Served through POST /v1/systemone at the provider's list price; input tokens only, output is free. Aliases jev-latest and jev-preview resolve to it.","tags":"system-one,typed-judgments,classification,routing,scoring,64K context,list price","vendor_id":6,"quota_type":0,"owner_by":"","enable_groups":["typesafe-official"],"supported_endpoint_types":["systemone"],"auto_route":"typesafe-official","context_window":64000},{"base_price":{"unit":"per_million_tokens","input":3,"output":15,"cache_read":0.3,"cache_write":3,"image_input":3,"image_output":15,"audio_input":3,"audio_output":15},"route_price_sources":{"kimi-official":"override"},"reference_price":{"unit":"per_million_tokens","input":3,"output":15,"cache_read":0.3,"cache_write":3},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://platform.kimi.ai/docs/pricing/chat","verified_at":"2026-09-24"},"route_prices":{"kimi-official":{"unit":"per_million_tokens","input":2.4,"output":12,"cache_read":0.24,"cache_write":2.4,"image_input":2.4,"image_output":12,"audio_input":2.4,"audio_output":12}},"model_name":"kimi-k3","reasoning":{"efforts":["low","high","max"],"default_effort":"max","wire_format":"chat"},"description":"Kimi K3 by Moonshot AI: 1M-token context model for coding and agentic work, served on one official line at 80% of the Moonshot list price.","tags":"chat,tools,reasoning,1M context,128K output","vendor_id":5,"quota_type":0,"owner_by":"","enable_groups":["kimi-official"],"supported_endpoint_types":["openai"],"auto_route":"kimi-official","context_window":1048576,"max_output":131072}],"pricing_version":"route-usd-v1","success":true,"supported_endpoint":{"anthropic":{"path":"/v1/messages","method":"POST"},"image-edit":{"path":"/v1/images/edits","method":"POST"},"image-generation":{"path":"/v1/images/generations","method":"POST"},"openai":{"path":"/v1/chat/completions","method":"POST"},"openai-response":{"path":"/v1/responses","method":"POST"},"systemone":{"path":"/v1/systemone","method":"POST"},"video-generation":{"path":"/v1/videos/generations","method":"POST"}},"time_tiers":[{"name":"deepseek","peak_windows_utc":["01:00-04:00","06:00-10:00"],"weekdays_only":true,"holidays_off_peak":true,"current_tier":"off_peak","next_change_at":1791421200,"holiday_coverage_end":"2026-10-07","holiday_calendar":"Asia/Shanghai"}],"usable_group":{"auto":"Auto route · Matches the model's group and routes to the lowest-priced line","claude-official":"Claude · Official key · 90% of list price","claude-shared":"Claude · Shared pool · 50% of list price","deepseek-official":"DeepSeek · Official key · 80% of list price · off-peak and peak hours","grok-official":"Grok · Official key · 90% of list price · video per second","grok-shared":"Grok · Shared pool · 50% of list price","image":"Image · Per image by size","kimi-official":"Kimi · Official key · 80% of list price","openai-official":"OpenAI · Official key · 90% of list price","openai-shared":"OpenAI · Shared pool · 50% of list price","typesafe-official":"TypeSafe · Official key · 107% of list price · System One","z-official":"GLM · Private deployment · fine-tuned, above list price"},"vendors":[{"id":1,"name":"Anthropic","description":"Claude models by Anthropic","icon":"Claude.Color"},{"id":2,"name":"OpenAI","description":"GPT models by OpenAI","icon":"OpenAI"},{"id":3,"name":"xAI","description":"Grok models by xAI","icon":"XAI"},{"id":4,"name":"Zhipu AI","description":"GLM models by Zhipu AI (Z.ai)","icon":"Zhipu"},{"id":5,"name":"Moonshot AI","description":"Kimi models by Moonshot AI"},{"id":6,"name":"TypeSafe","description":"System One models (Jev) by TypeSafe"},{"id":7,"name":"DeepSeek","description":"DeepSeek V4 models by DeepSeek"}]}