{"currency":"USD","data":[{"bestFor":"Trying MiniMax M3 on text, images, and video before paying","cachedInputPerMillionTokens":0,"contextTokens":1048576,"id":"minimax-m3-free","inputPerMillionTokens":0,"name":"MiniMax M3","outputPerMillionTokens":0,"serving":false,"supports":["text","reasoning","tool-use","image input","vision","video input"]},{"bestFor":"Low-cost text agents with long prompts and reusable prefixes","cachedInputPerMillionTokens":0.02,"contextTokens":1048576,"id":"dsv4","inputPerMillionTokens":0.04,"name":"DeepSeek V4 Flash","outputPerMillionTokens":0.21,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"Low-cost omnimodal work across text, images, audio, and video","cachedInputPerMillionTokens":0.0028,"contextTokens":1048576,"id":"mimo-v2.6-flash","inputPerMillionTokens":0.14,"name":"MiMo-V2.6 Flash","outputPerMillionTokens":0.28,"serving":true,"supports":["text","reasoning","tool-use","image input","vision","audio input","video input"]},{"bestFor":"Affordable image and text tasks with a shorter context","cachedInputPerMillionTokens":0.02,"contextTokens":262144,"id":"qwen3.6","inputPerMillionTokens":0.04,"name":"Qwen3.6-35B-A3B","outputPerMillionTokens":0.35,"serving":true,"supports":["text","reasoning","tool-use","image input","vision"]},{"bestFor":"Lower-cost coding and visual tasks with a million-token context","cachedInputPerMillionTokens":0.024,"contextTokens":1048576,"id":"glm-5.3-flash","inputPerMillionTokens":0.12,"name":"GLM-5.3 Flash","outputPerMillionTokens":0.4,"serving":true,"supports":["text","reasoning","tool-use","image input","vision","video input"]},{"bestFor":"Quick image analysis with a million-token context","cachedInputPerMillionTokens":0.016,"contextTokens":1000000,"id":"qwen3.8-flash","inputPerMillionTokens":0.15,"name":"Qwen3.8 Flash","outputPerMillionTokens":0.47,"serving":true,"supports":["text","reasoning","tool-use","image input","vision"]},{"bestFor":"Everyday GPT-6 work where you want to keep costs down","cachedInputPerMillionTokens":0.01,"contextTokens":1050000,"id":"gpt-6-luna","inputPerMillionTokens":0.1,"name":"GPT-6 Luna","outputPerMillionTokens":0.5,"serving":true,"supports":["text","reasoning","tool-use","image input","vision"]},{"bestFor":"Coordinating agents, planning work, and combining their results","cachedInputPerMillionTokens":0.06,"contextTokens":262144,"id":"nemotron3-super","inputPerMillionTokens":0.12,"name":"Nemotron 3 Super","outputPerMillionTokens":0.52,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"A frontier open-weight agent that reads text, images, audio, and video","cachedInputPerMillionTokens":0.0036,"contextTokens":1048576,"id":"mimo-v2.6-pro","inputPerMillionTokens":0.435,"name":"MiMo-V2.6 Pro","outputPerMillionTokens":0.87,"serving":true,"supports":["text","reasoning","tool-use","image input","vision","audio input","video input"]},{"bestFor":"Fast coding agents that need image input and a long context","cachedInputPerMillionTokens":0.01,"contextTokens":1048576,"id":"deepseek-v4.1-flash","inputPerMillionTokens":0.2,"name":"DeepSeek V4.1 Flash","outputPerMillionTokens":1,"serving":true,"supports":["text","reasoning","tool-use","image input","vision"]},{"bestFor":"Long-running coding agents that work with images and video","cachedInputPerMillionTokens":0.056,"contextTokens":1048576,"id":"minimax-m3","inputPerMillionTokens":0.28,"name":"MiniMax M3","outputPerMillionTokens":1.1,"serving":true,"supports":["text","reasoning","tool-use","image input","vision","video input"]},{"bestFor":"High-volume GPT-5.6 chat, classification, and light agents","cachedInputPerMillionTokens":0.02,"contextTokens":262144,"id":"gpt-5.6-luna","inputPerMillionTokens":0.2,"name":"GPT-5.6 Luna","outputPerMillionTokens":1.2,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"A main coding agent paired with DeepSeek V4 Flash sub-agents","cachedInputPerMillionTokens":0.0352,"contextTokens":1048576,"id":"deepseek-v4-pro-0813","inputPerMillionTokens":1.04,"name":"DeepSeek V4 Pro","outputPerMillionTokens":2.08,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"Long coding tasks that run for hours and work with text","cachedInputPerMillionTokens":0.17,"contextTokens":1048576,"id":"glm-5.2","inputPerMillionTokens":0.34,"name":"GLM-5.2","outputPerMillionTokens":2.3,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"Long-running coding agents that need a million-token context","cachedInputPerMillionTokens":0.14,"contextTokens":1048576,"id":"glm-5.3","inputPerMillionTokens":1.12,"name":"GLM-5.3","outputPerMillionTokens":3.52,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"GPT-5.6 flagship for CLI-heavy coding and long-horizon problem solving","cachedInputPerMillionTokens":0.2,"contextTokens":262144,"id":"gpt-5.6-sol","inputPerMillionTokens":2,"name":"GPT-5.6 Sol","outputPerMillionTokens":10,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"Harder coding problems and agents that need to work for longer","cachedInputPerMillionTokens":0.2,"contextTokens":1050000,"id":"gpt-6-sol","inputPerMillionTokens":2,"name":"GPT-6 Sol","outputPerMillionTokens":10,"serving":true,"supports":["text","reasoning","tool-use","image input","vision"]},{"bestFor":"Complex coding and reasoning across long-running tasks","cachedInputPerMillionTokens":0.23,"contextTokens":1048576,"id":"kimi-k3","inputPerMillionTokens":2.3,"name":"Kimi K3","outputPerMillionTokens":11.4,"serving":true,"supports":["text","reasoning","tool-use","image input","vision"]},{"bestFor":"Long coding or research tasks that benefit from adaptive thinking","cachedInputPerMillionTokens":0.2,"contextTokens":1000000,"id":"claude-opus-5.5","inputPerMillionTokens":4,"name":"Claude Opus 5.5","outputPerMillionTokens":20,"serving":true,"supports":["text","reasoning","tool-use","image input","vision"]},{"bestFor":"Anthropic Opus 5 for demanding coding, review, and long-horizon agents","cachedInputPerMillionTokens":0.5,"contextTokens":262144,"id":"claude-opus-5","inputPerMillionTokens":5,"name":"Claude Opus 5","outputPerMillionTokens":25,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"Coding agents, front-end work, and research-heavy tasks","cachedInputPerMillionTokens":0.25,"contextTokens":262144,"id":"claude-fable-5.1","inputPerMillionTokens":10,"name":"Claude Fable 5.1","outputPerMillionTokens":50,"serving":true,"supports":["text","reasoning","tool-use"]},{"bestFor":"Routing, classification, and guardrail decisions with calibrated probabilities","cachedInputPerMillionTokens":null,"contextTokens":8192,"id":"convaiinnovations/laya","inputPerMillionTokens":null,"name":"Laya","outputPerMillionTokens":null,"serving":true,"supports":["text"]}],"unit":"per_million_tokens"}