{"data":[{"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":"2026-11-08","model":"GLM-5.3"},"route_prices":{"z-official":{"unit":"per_million_tokens","input":3,"output":6,"cache_read":0.45,"image_input":3,"image_output":6,"audio_input":3,"audio_output":6,"backup":{"input":4,"output":7,"cache_read":0.4}}},"model_name":"GLM-5.3-Abliterated","reasoning":{"efforts":["low","high","max"],"default_effort":"low","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 (the default), high and max.","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},{"route_price_sources":{"z-official":"override"},"reference_price":{"unit":"per_million_tokens","input":0.15,"output":0.5,"cache_read":0.03},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://docs.z.ai/guides/overview/pricing","verified_at":"2026-11-08","model":"GLM-5.3-Flash"},"route_prices":{"z-official":{"unit":"per_million_tokens","input":0.36,"output":1.5,"cache_read":0.075,"image_input":0.36,"image_output":1.5,"audio_input":0.36,"audio_output":1.5}},"model_name":"GLM-5.3-Flash-Abliterated","reasoning":{"efforts":["low","high","max"],"default_effort":"low","wire_format":"chat"},"description":"GLM-5.3-Flash-Abliterated: Z.ai's GLM-5.3-Flash with the built-in refusal alignment removed (abliterated), served from a private deployment, not the provider's official API. Text and image input with a 1,048,576-token context, for fast, low-cost work; it always reasons, in three modes: low (the default), high and max.","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,vision,tools,reasoning,1M context,128K output,abliterated,private deployment","vendor_id":4,"quota_type":0,"owner_by":"","enable_groups":["z-official"],"supported_endpoint_types":["openai","openai-response"],"auto_route":"z-official","context_window":1048576,"max_output":131072},{"base_price":{"unit":"per_million_tokens","input":8,"output":40,"cache_read":0.2,"cache_write":10,"cache_write_1h":16,"image_input":8,"image_output":40,"audio_input":8,"audio_output":40},"route_price_sources":{"claude-official":"override","claude-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":8,"output":40,"cache_read":0.2,"cache_write":10,"cache_write_1h":16},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://platform.claude.com/docs/en/about-claude/pricing","verified_at":"2026-11-10"},"route_prices":{"claude-official":{"unit":"per_million_tokens","input":7.2,"output":36,"cache_read":0.18,"cache_write":9,"cache_write_1h":14.4,"image_input":7.2,"image_output":36,"audio_input":7.2,"audio_output":36},"claude-shared":{"unit":"per_million_tokens","input":4,"output":20,"cache_read":0.1,"cache_write":5,"cache_write_1h":8,"image_input":4,"image_output":20,"audio_input":4,"audio_output":20}},"model_name":"claude-fable-5-2","reasoning":{"efforts":["low","medium","high","xhigh","max"],"default_effort":"high","wire_format":"chat"},"description":"Claude Fable 5.2: Anthropic's most capable widely released model for demanding reasoning, coding and long agent runs, 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":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":"2026-10-24"},"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":0.1,"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":0.1,"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":"2026-10-24"},"route_prices":{"claude-official":{"unit":"per_million_tokens","input":1.8,"output":9,"cache_read":0.09,"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":0.05,"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},{"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":"2026-11-08","model":"deepseek-flash"},"route_prices":{"deepseek-official":{"unit":"per_million_tokens","input":0.42,"output":1.5,"cache_read":0.105,"image_input":0.42,"image_output":1.5,"audio_input":0.42,"audio_output":1.5,"backup":{"input":1.3,"output":2.6,"cache_read":0.15}}},"model_name":"deepseek-flash-Abliterated","reasoning":{"efforts":["none","low","high","max"],"default_effort":"low","wire_format":"chat"},"description":"deepseek-flash-Abliterated: DeepSeek V4.1 Flash 1017 with the built-in refusal alignment removed (abliterated), from a private deployment, supporting text and images, with a 1,048,576-token context and up to 131,072 output tokens. Reasoning modes are none, low (the default), high and max; none turns reasoning off. Fixed prices apply at all hours, without DeepSeek's peak-hour multiplier.","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,vision,reasoning,1M context,128K output,abliterated,private deployment","vendor_id":7,"quota_type":0,"owner_by":"","enable_groups":["deepseek-official"],"supported_endpoint_types":["openai","openai-response"],"auto_route":"deepseek-official","context_window":1048576,"max_output":131072},{"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":"2026-11-08","model":"deepseek-flash"},"route_prices":{"deepseek-official":{"unit":"per_million_tokens","input":0.02,"output":0.16,"cache_read":0.004,"image_input":0.02,"image_output":0.16,"audio_input":0.02,"audio_output":0.16}},"model_name":"deepseek-flash-fast","reasoning":{"efforts":["none","low","high","max"],"default_effort":"low","wire_format":"chat"},"description":"deepseek-flash-fast: DeepSeek V4.1 Flash 1017, the current version, on its standard weights from a private deployment. It is not abliterated and keeps the base model's refusal behaviour. Text and image input, a 1,048,576-token context and up to 262,144 output tokens. Reasoning modes are none, low (the default), high and max; none turns reasoning off. Fixed prices apply at all hours, without DeepSeek's peak-hour multiplier.","tags":"chat,tools,vision,reasoning,1M context,256K output,private deployment","vendor_id":7,"quota_type":0,"owner_by":"","enable_groups":["deepseek-official"],"supported_endpoint_types":["openai","openai-response"],"auto_route":"deepseek-official","context_window":1048576,"max_output":262144},{"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":"2026-10-24","note":"Cache write references use the default 30-minute TTL."},"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"},"description":"OpenAI's flagship GPT-6 Astra: top-tier reasoning and coding 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.","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_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":"2026-10-24","note":"Cache write references use the default 30-minute TTL."},"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_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}},"route_price_sources":{"openai-official":"override","openai-shared":"override"},"reference_price":{"unit":"per_million_tokens","input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"long_context":{"min_input_tokens":272001,"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://developers.openai.com/api/docs/models/gpt-6.1-terra","verified_at":"2026-10-30","note":"Cache write references use the default 30-minute TTL."},"route_prices":{"openai-official":{"unit":"per_million_tokens","input":0.9,"output":4.5,"cache_read":0.045,"cache_write":1.125,"image_input":0.9,"image_output":4.5,"audio_input":0.9,"audio_output":4.5,"long_context":{"min_input_tokens":272001,"input":1.8,"output":6.75,"cache_read":0.09,"cache_write":2.25,"image_input":1.8,"image_output":6.75,"audio_input":1.8,"audio_output":6.75}},"openai-shared":{"unit":"per_million_tokens","input":0.5,"output":2.5,"cache_read":0.025,"cache_write":0.625,"image_input":0.5,"image_output":2.5,"audio_input":0.5,"audio_output":2.5,"long_context":{"min_input_tokens":272001,"input":1,"output":3.75,"cache_read":0.05,"cache_write":1.25,"image_input":1,"image_output":3.75,"audio_input":1,"audio_output":3.75}}},"model_name":"gpt-6.1-terra","reasoning":{"efforts":["none","low","medium","high","xhigh","max"],"default_effort":"medium","wire_format":"responses"},"description":"GPT-6.1 Terra: fast, efficient visual reasoning, computer use and writing 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 can be turned off; tool calling is 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,"4k":0.2}},"route_price_sources":{"image":"override"},"route_prices":{"image":{"unit":"per_image","sizes":{"1k":0.03,"2k":0.05,"4k":0.2}}},"model_name":"gpt-image-2.6-flare","reasoning":null,"description":"GPT Image 2.6 Flare: OpenAI's image generation and editing model for fast iteration, sold per image by output size (1K, 2K or 4K).","tags":"image,image-generation,image-edit,per image,1K,2K,4K","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,"4k":0.2}},"route_price_sources":{"image":"override"},"route_prices":{"image":{"unit":"per_image","sizes":{"1k":0.03,"2k":0.05,"4k":0.2}}},"model_name":"gpt-image-2.6-sunburst","reasoning":null,"description":"GPT Image 2.6 Sunburst: OpenAI's image generation and editing model for detailed visual work, sold per image by output size (1K, 2K or 4K).","tags":"image,image-generation,image-edit,per image,1K,2K,4K","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":"2026-10-24"},"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","resolutions":{"1080p":0.25,"480p":0.08,"720p":0.14}},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://docs.x.ai/developers/models/grok-imagine-video-1.5","verified_at":"2026-10-24","note":"Output generation prices only. The provider also charges $0.01 per input image."},"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},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://docs.typesafe.ai/models","verified_at":"2026-11-12"},"route_prices":{"typesafe-official":{"unit":"per_million_tokens","input":0.055,"output":0,"cache_read":0.055,"image_input":0.055,"image_output":0,"audio_input":0.055,"audio_output":0,"backup":{"input":0.058,"output":0,"cache_read":0.058}}},"model_name":"jev-1.14.0","reasoning":null,"description":"Jev 1.14.0 by TypeSafe, the System One model: it answers typed questions (noul, choice, score) about a state with calibrated probabilities instead of generating text, and can look at PNG or JPEG images placed in the state, instructions or criteria. Box questions return where something is in an image. Images and Box are an experimental TypeSafe feature. Served through POST /v1/systemone at a fixed price; input tokens only (image tokens included), output is free. A costlier standby line answers, rarely, when the main line fails, at the backup price. jev-latest and jev-preview are answered by it.","tags":"system-one,typed-judgments,classification,routing,scoring,images,box,64K context","vendor_id":6,"quota_type":0,"owner_by":"","enable_groups":["typesafe-official"],"supported_endpoint_types":["systemone"],"auto_route":"typesafe-official","context_window":64000},{"route_price_sources":{"kimi-official":"override"},"reference_price":{"unit":"per_million_tokens","input":2,"output":10,"cache_read":0.2},"reference_price_source":{"kind":"provider_official","name":"Provider published price","url":"https://platform.kimi.ai/docs/pricing/chat","verified_at":"2026-11-08"},"route_prices":{"kimi-official":{"unit":"per_million_tokens","input":0.5,"output":5,"cache_read":0.3,"image_input":0.5,"image_output":5,"audio_input":0.5,"audio_output":5,"backup":{"input":2.4,"output":12,"cache_read":0.24,"cache_write":2.4}}},"model_name":"kimi-k3.1-pro","reasoning":{"efforts":["low","high","max"],"default_effort":"low","wire_format":"chat"},"description":"Kimi K3.1 Pro by Moonshot AI, served from a private deployment rather than Moonshot's own API: text and image input with a 1,048,576-token context for coding and agentic work. Reasoning is always enabled, with low, high and max modes; low is the default.","tags":"chat,vision,tools,reasoning,1M context,128K output,private deployment","vendor_id":5,"quota_type":0,"owner_by":"","enable_groups":["kimi-official"],"supported_endpoint_types":["openai","openai-response"],"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":"peak","next_change_at":1794542400,"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","claude-special":"Claude · Special offer · Opus 5.5 only · limited time, often unavailable","deepseek-official":"DeepSeek · Private deployment · fixed prices","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 · Private deployment · fixed prices","openai-official":"OpenAI · Official key · 90% of list price","openai-shared":"OpenAI · Shared pool · 50% of list price","typesafe-official":"TypeSafe · Official key · fixed price · System One","z-official":"GLM · Private deployment · fixed prices"},"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"}]}