{"ok":true,"license":"CC-BY-4.0","builtAt":"2026-11-06T08:41:19.516Z","source":"https://openrouter.ai/api/v1/models","models":[{"id":"aion-labs/aion-2.0","name":"Aion-2.0","vendor":"aion-labs","releasedAt":"2026-02-23","contextLength":131072,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling. It is particularly strong at introducing tension, crises, and conflict into stories, making narratives feel more engaging....","inputPerM":0.8,"outputPerM":1.6,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.8,"outputPerM":1.6,"cacheReadPerM":0.2}]},{"id":"aion-labs/aion-3.0","name":"Aion-3.0","vendor":"aion-labs","releasedAt":"2026-07-07","contextLength":131072,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Aion-3.0 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each contribute...","inputPerM":3,"outputPerM":6,"cacheReadPerM":0.75,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":6,"cacheReadPerM":0.75}]},{"id":"aion-labs/aion-3.0-mini","name":"Aion-3.0-Mini","vendor":"aion-labs","releasedAt":"2026-07-07","contextLength":131072,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Aion-3.0 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the DeepSeek family of models. It uses a collaborative generation process in which multiple specialized models each...","inputPerM":0.7,"outputPerM":1.4,"cacheReadPerM":0.18,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.7,"outputPerM":1.4,"cacheReadPerM":0.18}]},{"id":"aion-labs/aion-3.5","name":"Aion 3.5","vendor":"aion-labs","releasedAt":"2026-09-23","contextLength":262144,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Aion 3.5 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each...","inputPerM":3,"outputPerM":6,"cacheReadPerM":0.75,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":6,"cacheReadPerM":0.75}]},{"id":"aion-labs/aion-3.5-mini","name":"Aion 3.5 Mini","vendor":"aion-labs","releasedAt":"2026-09-23","contextLength":262144,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Aion 3.5 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It is the smaller, lower-cost sibling of Aion 3.5 and uses...","inputPerM":0.7,"outputPerM":1.4,"cacheReadPerM":0.18,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.7,"outputPerM":1.4,"cacheReadPerM":0.18}]},{"id":"aion-labs/aion-rp-llama-3.1-8b","name":"Aion-RP 1.0 (8B)","vendor":"aion-labs","releasedAt":"2025-02-04","contextLength":32768,"maxOutput":29491,"input":["text"],"output":["text"],"features":[],"description":"Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses. It is a fine-tuned base model...","inputPerM":0.8,"outputPerM":1.6,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.8,"outputPerM":1.6,"cacheReadPerM":null}]},{"id":"amazon/nova-2-lite-v1","name":"Nova 2 Lite","vendor":"amazon","releasedAt":"2025-12-02","contextLength":1000000,"maxOutput":65535,"input":["text","image","video","file"],"output":["text"],"features":["reasoning","tools"],"description":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text. Nova 2 Lite demonstrates standout capabilities in processing...","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":null}]},{"id":"amazon/nova-lite-v1","name":"Nova Lite 1.0","vendor":"amazon","releasedAt":"2024-12-05","contextLength":300000,"maxOutput":5120,"input":["text","image"],"output":["text"],"features":["tools"],"description":"Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output. Amazon Nova Lite...","inputPerM":0.06,"outputPerM":0.24,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.06,"outputPerM":0.24,"cacheReadPerM":null}]},{"id":"amazon/nova-micro-v1","name":"Nova Micro 1.0","vendor":"amazon","releasedAt":"2024-12-05","contextLength":128000,"maxOutput":5120,"input":["text"],"output":["text"],"features":["tools"],"description":"Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost. With a context length...","inputPerM":0.035,"outputPerM":0.14,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.035,"outputPerM":0.14,"cacheReadPerM":null}]},{"id":"amazon/nova-premier-v1","name":"Nova Premier 1.0","vendor":"amazon","releasedAt":"2025-10-31","contextLength":1000000,"maxOutput":32000,"input":["text","image"],"output":["text"],"features":["tools"],"description":"Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.","inputPerM":2.5,"outputPerM":12.5,"cacheReadPerM":0.625,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":12.5,"cacheReadPerM":0.625}]},{"id":"amazon/nova-pro-v1","name":"Nova Pro 1.0","vendor":"amazon","releasedAt":"2024-12-05","contextLength":300000,"maxOutput":5120,"input":["text","image"],"output":["text"],"features":["tools"],"description":"Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks. As of December...","inputPerM":0.8,"outputPerM":3.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.8,"outputPerM":3.2,"cacheReadPerM":null}]},{"id":"anthracite-org/magnum-v4-72b","name":"Magnum v4 72B","vendor":"anthracite-org","releasedAt":"2024-10-22","contextLength":32768,"maxOutput":4096,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus).\n\nThe model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-2.5-72b-instruct).","inputPerM":2.5,"outputPerM":5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":5,"cacheReadPerM":null}]},{"id":"anthropic/claude-fable-5","name":"Claude Fable 5","vendor":"anthropic","releasedAt":"2026-06-09","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...","inputPerM":10,"outputPerM":50,"cacheReadPerM":1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":10,"outputPerM":50,"cacheReadPerM":1}]},{"id":"anthropic/claude-fable-5.1","name":"Claude Fable 5.1","vendor":"anthropic","releasedAt":"2026-09-01","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual...","inputPerM":10,"outputPerM":50,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":10,"outputPerM":50,"cacheReadPerM":0.25}]},{"id":"anthropic/claude-haiku-4.5","name":"Claude Haiku 4.5","vendor":"anthropic","releasedAt":"2025-10-15","contextLength":200000,"maxOutput":64000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...","inputPerM":1,"outputPerM":5,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":5,"cacheReadPerM":0.1}]},{"id":"anthropic/claude-haiku-5.5","name":"Claude Haiku 5.5","vendor":"anthropic","releasedAt":"2026-10-07","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...","inputPerM":0.1,"outputPerM":0.5,"cacheReadPerM":0.01,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.5,"cacheReadPerM":0.01}]},{"id":"anthropic/claude-opus-4.1","name":"Claude Opus 4.1","vendor":"anthropic","releasedAt":"2025-08-05","contextLength":200000,"maxOutput":32000,"input":["image","text","file"],"output":["text"],"features":["reasoning","tools"],"description":"Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks. It achieves 74.5% on SWE-bench Verified and shows notable gains...","inputPerM":15,"outputPerM":75,"cacheReadPerM":1.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":15,"outputPerM":75,"cacheReadPerM":1.5}]},{"id":"anthropic/claude-opus-4.5","name":"Claude Opus 4.5","vendor":"anthropic","releasedAt":"2025-11-24","contextLength":200000,"maxOutput":64000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use. It offers strong multimodal capabilities, competitive performance across real-world coding and...","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5}]},{"id":"anthropic/claude-opus-4.6","name":"Claude Opus 4.6","vendor":"anthropic","releasedAt":"2026-02-04","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective...","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5}]},{"id":"anthropic/claude-opus-4.7","name":"Claude Opus 4.7","vendor":"anthropic","releasedAt":"2026-04-16","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on...","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5}]},{"id":"anthropic/claude-opus-4.8","name":"Claude Opus 4.8","vendor":"anthropic","releasedAt":"2026-05-27","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5}]},{"id":"anthropic/claude-opus-5","name":"Claude Opus 5","vendor":"anthropic","releasedAt":"2026-07-24","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":25,"cacheReadPerM":0.5}]},{"id":"anthropic/claude-opus-5.5","name":"Claude Opus 5.5","vendor":"anthropic","releasedAt":"2026-09-22","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.2}]},{"id":"anthropic/claude-sonnet-4.5","name":"Claude Sonnet 4.5","vendor":"anthropic","releasedAt":"2025-09-29","contextLength":1000000,"maxOutput":64000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with...","inputPerM":3,"outputPerM":15,"cacheReadPerM":0.3,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":15,"cacheReadPerM":0.3}]},{"id":"anthropic/claude-sonnet-4.6","name":"Claude Sonnet 4.6","vendor":"anthropic","releasedAt":"2026-02-17","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with...","inputPerM":3,"outputPerM":15,"cacheReadPerM":0.3,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":15,"cacheReadPerM":0.3}]},{"id":"anthropic/claude-sonnet-5","name":"Claude Sonnet 5","vendor":"anthropic","releasedAt":"2026-06-30","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2}]},{"id":"anthropic/claude-sonnet-5.5","name":"Claude Sonnet 5.5","vendor":"anthropic","releasedAt":"2026-09-28","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1}]},{"id":"bytedance/ui-tars-1.5-7b","name":"UI-TARS 7B","vendor":"bytedance","releasedAt":"2025-07-22","contextLength":128000,"maxOutput":2048,"input":["image","text"],"output":["text"],"features":["structured_outputs"],"description":"UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games. Built by ByteDance, it builds upon the UI-TARS framework with reinforcement...","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":0.1}]},{"id":"bytedance-seed/seed-1.6","name":"Seed 1.6","vendor":"bytedance-seed","releasedAt":"2025-12-23","contextLength":262144,"maxOutput":32768,"input":["image","text","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Seed 1.6 is a general-purpose model released by the ByteDance Seed team. It incorporates multimodal capabilities and adaptive deep thinking with a 256K context window.","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":null}]},{"id":"bytedance-seed/seed-1.6-flash","name":"Seed 1.6 Flash","vendor":"bytedance-seed","releasedAt":"2025-12-23","contextLength":262144,"maxOutput":32768,"input":["image","text","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Seed 1.6 Flash is an ultra-fast multimodal deep thinking model by ByteDance Seed, supporting both text and visual understanding. It features a 256k context window and can generate outputs of...","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":null}]},{"id":"bytedance-seed/seed-2.0-code","name":"Seed-2.0-Code","vendor":"bytedance-seed","releasedAt":"2026-08-12","contextLength":262144,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Seed 2.0 Code is a model from ByteDance Seed optimized for agentic coding. It is suited for frontend development, multilingual programming tasks, and coding-agent workflows in tools such as Claude...","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":null}]},{"id":"bytedance-seed/seed-2.0-lite","name":"Seed-2.0-Lite","vendor":"bytedance-seed","releasedAt":"2026-03-10","contextLength":262144,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably lower latency, making it a practical default choice for most production workloads across...","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":null}]},{"id":"bytedance-seed/seed-2.0-mini","name":"Seed-2.0-Mini","vendor":"bytedance-seed","releasedAt":"2026-02-26","contextLength":262144,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment. It delivers performance comparable to ByteDance-Seed-1.6, supports 256k context, four reasoning effort modes (minimal/low/medium/high), multimodal understanding,...","inputPerM":0.1,"outputPerM":0.4,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.4,"cacheReadPerM":null}]},{"id":"bytedance-seed/seed-2-1-turbo","name":"Seed 2.1 Turbo","vendor":"bytedance-seed","releasedAt":"2026-08-12","contextLength":262144,"maxOutput":235929,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Seed 2.1 Turbo is a multimodal model from ByteDance Seed for coding and long-horizon agent workflows. It is suited for end-to-end software delivery, multi-step task execution, and understanding visual and...","inputPerM":0.5,"outputPerM":2.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":2.5,"cacheReadPerM":null}]},{"id":"cognitivecomputations/dolphin-mistral-24b-venice-edition","name":"Uncensored","vendor":"cognitivecomputations","releasedAt":"2025-07-09","contextLength":128000,"maxOutput":8192,"input":["text"],"output":["text"],"features":[],"description":"Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai. This model is designed as an “uncensored” instruct-tuned LLM, preserving...","inputPerM":0.2,"outputPerM":0.9,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":0.9,"cacheReadPerM":null}]},{"id":"cohere/command-a","name":"Command A","vendor":"cohere","releasedAt":"2025-03-13","contextLength":256000,"maxOutput":8192,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases. Compared to other leading proprietary...","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":null}]},{"id":"cohere/command-a-plus","name":"Command A+","vendor":"cohere","releasedAt":"2026-09-22","contextLength":192000,"maxOutput":64000,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Command A+ is Cohere's flagship model for enterprise agentic workflows. It accepts text and image inputs with a 192K context window, supports native tool calling with strict tool schemas, structured...","inputPerM":0.3,"outputPerM":1.5,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":1.5,"cacheReadPerM":0.15}]},{"id":"cohere/command-r-08-2024","name":"Command R (08-2024)","vendor":"cohere","releasedAt":"2024-08-30","contextLength":128000,"maxOutput":4000,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use. More broadly, it is better at math, code and reasoning and...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":null}]},{"id":"cohere/command-r7b-12-2024","name":"Command R7B (12-2024)","vendor":"cohere","releasedAt":"2024-12-14","contextLength":128000,"maxOutput":4000,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG, tool use, agents, and similar tasks requiring complex reasoning...","inputPerM":0.0375,"outputPerM":0.15,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.0375,"outputPerM":0.15,"cacheReadPerM":null}]},{"id":"cohere/command-r-plus-08-2024","name":"Command R+ (08-2024)","vendor":"cohere","releasedAt":"2024-08-30","contextLength":128000,"maxOutput":4000,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint...","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":null}]},{"id":"deepseek/deepseek-chat","name":"DeepSeek V3","vendor":"deepseek","releasedAt":"2024-12-26","contextLength":163840,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...","inputPerM":0.32,"outputPerM":0.89,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2574,"outputPerM":1.0287,"cacheReadPerM":null},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.32,"outputPerM":0.89,"cacheReadPerM":null}]},{"id":"deepseek/deepseek-chat-v3-0324","name":"DeepSeek V3 0324","vendor":"deepseek","releasedAt":"2025-03-24","contextLength":163840,"maxOutput":115200,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) model and performs really well...","inputPerM":0.29,"outputPerM":1.14,"cacheReadPerM":0.11,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.29,"outputPerM":1.14,"cacheReadPerM":0.11}]},{"id":"deepseek/deepseek-chat-v3.1","name":"DeepSeek V3.1","vendor":"deepseek","releasedAt":"2025-08-21","contextLength":163840,"maxOutput":147456,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...","inputPerM":0.25,"outputPerM":0.95,"cacheReadPerM":0.13,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":0.95,"cacheReadPerM":0.13}]},{"id":"deepseek/deepseek-r1-0528","name":"R1 0528","vendor":"deepseek","releasedAt":"2025-05-28","contextLength":163840,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active...","inputPerM":0.5,"outputPerM":2.15,"cacheReadPerM":0.35,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":2.15,"cacheReadPerM":0.35}]},{"id":"deepseek/deepseek-v3.1-terminus","name":"DeepSeek V3.1 Terminus","vendor":"deepseek","releasedAt":"2025-09-22","contextLength":163840,"maxOutput":147456,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's...","inputPerM":0.27,"outputPerM":1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.27,"outputPerM":1,"cacheReadPerM":null}]},{"id":"deepseek/deepseek-v3.2","name":"DeepSeek V3.2","vendor":"deepseek","releasedAt":"2025-12-01","contextLength":163840,"maxOutput":147456,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...","inputPerM":0.26,"outputPerM":0.38,"cacheReadPerM":0.13,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-30","inputPerM":0.28,"outputPerM":0.42,"cacheReadPerM":0.028},{"from":"2026-10-30","to":"2026-11-06","inputPerM":0.26,"outputPerM":0.38,"cacheReadPerM":0.13}]},{"id":"deepseek/deepseek-v3.2-exp","name":"DeepSeek V3.2 Exp","vendor":"deepseek","releasedAt":"2025-09-29","contextLength":163840,"maxOutput":147456,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...","inputPerM":0.27,"outputPerM":0.41,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.27,"outputPerM":0.41,"cacheReadPerM":null}]},{"id":"deepseek/deepseek-v4.1-flash","name":"DeepSeek V4.1 Flash","vendor":"deepseek","releasedAt":"2026-09-10","contextLength":1048576,"maxOutput":943718,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...","inputPerM":0.051,"outputPerM":0.567,"cacheReadPerM":0.018,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.14,"outputPerM":0.42,"cacheReadPerM":0.0042},{"from":"2026-10-29","to":"2026-11-03","inputPerM":0.045,"outputPerM":0.5,"cacheReadPerM":0.006},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.045,"outputPerM":0.5,"cacheReadPerM":0.012},{"from":"2026-11-04","to":"2026-11-06","inputPerM":0.045,"outputPerM":0.5,"cacheReadPerM":0.018},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.051,"outputPerM":0.567,"cacheReadPerM":0.018}]},{"id":"deepseek/deepseek-v4.1-flash-1017","name":"DeepSeek V4.1 Flash 1017","vendor":"deepseek","releasedAt":"2026-10-17","contextLength":1048576,"maxOutput":393216,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name. V4.1 Flash is a sparse mixture-of-experts model...","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-29","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.003},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006},{"from":"2026-10-29","to":"2026-10-30","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.003},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006},{"from":"2026-10-30","to":"2026-11-01","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.003},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.003},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.003},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.003},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.003},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006}]},{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek V4 Flash 0423","vendor":"deepseek","releasedAt":"2026-04-24","contextLength":1048576,"maxOutput":943717,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...","inputPerM":0.091,"outputPerM":0.182,"cacheReadPerM":0.0182,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.091,"outputPerM":0.182,"cacheReadPerM":0.0182}]},{"id":"deepseek/deepseek-v4-flash-0731","name":"DeepSeek V4 Flash 0731","vendor":"deepseek","releasedAt":"2026-07-31","contextLength":1048576,"maxOutput":943718,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....","inputPerM":0.059,"outputPerM":0.3,"cacheReadPerM":0.035,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-02","inputPerM":0.119,"outputPerM":0.238,"cacheReadPerM":0.0238},{"from":"2026-11-03","to":"2026-11-04","inputPerM":0.059,"outputPerM":0.3,"cacheReadPerM":0.035},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.0487,"outputPerM":0.3,"cacheReadPerM":0.035},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.059,"outputPerM":0.3,"cacheReadPerM":0.035},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.054,"outputPerM":0.3,"cacheReadPerM":0.035},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.0434,"outputPerM":0.3,"cacheReadPerM":0.035},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.059,"outputPerM":0.3,"cacheReadPerM":0.035}]},{"id":"deepseek/deepseek-v4-flash-vision-exp","name":"DeepSeek V4 Flash Vision Exp","vendor":"deepseek","releasedAt":"2026-08-21","contextLength":1048576,"maxOutput":943717,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of [DeepSeek V4 Flash 0731](https://openrouter.ai/deepseek/deepseek-v4-flash-0731) from DeepSeek, adding image understanding while matching the base model on text capabilities including agents,...","inputPerM":0.44,"outputPerM":1.32,"cacheReadPerM":0.014,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-05","inputPerM":0.2156,"outputPerM":0.6468,"cacheReadPerM":0.00686},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.44,"outputPerM":1.32,"cacheReadPerM":0.014}]},{"id":"deepseek/deepseek-v4-pro","name":"DeepSeek V4 Pro 0423","vendor":"deepseek","releasedAt":"2026-04-24","contextLength":1048576,"maxOutput":393216,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...","inputPerM":1.6,"outputPerM":3.2,"cacheReadPerM":0.135,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.95526,"outputPerM":1.91052,"cacheReadPerM":0.079605},{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.888966,"outputPerM":1.777932,"cacheReadPerM":0.074081},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.95526,"outputPerM":1.91052,"cacheReadPerM":0.079605},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.897318,"outputPerM":1.794636,"cacheReadPerM":0.074777},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.863562,"outputPerM":1.727124,"cacheReadPerM":0.071964},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.816234,"outputPerM":1.632468,"cacheReadPerM":0.06802},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.9483,"outputPerM":1.8966,"cacheReadPerM":0.079025},{"from":"2026-10-31","to":"2026-11-05","inputPerM":1.6,"outputPerM":3.2,"cacheReadPerM":0.135},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.95526,"outputPerM":1.91052,"cacheReadPerM":0.079605},{"from":"2026-11-06","to":"2026-11-06","inputPerM":1.6,"outputPerM":3.2,"cacheReadPerM":0.135}]},{"id":"deepseek/deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","vendor":"deepseek","releasedAt":"2026-08-12","contextLength":1048576,"maxOutput":943718,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.","inputPerM":0.1165,"outputPerM":4.66,"cacheReadPerM":0.111,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-29","inputPerM":0.66,"outputPerM":1.98,"cacheReadPerM":0.022},{"from":"2026-10-29","to":"2026-10-29","inputPerM":1.32,"outputPerM":3.96,"cacheReadPerM":0.044},{"from":"2026-10-29","to":"2026-10-30","inputPerM":0.66,"outputPerM":1.98,"cacheReadPerM":0.022},{"from":"2026-10-30","to":"2026-10-30","inputPerM":1.32,"outputPerM":3.96,"cacheReadPerM":0.044},{"from":"2026-10-30","to":"2026-10-31","inputPerM":0.66,"outputPerM":1.98,"cacheReadPerM":0.022},{"from":"2026-10-31","to":"2026-11-04","inputPerM":0.115,"outputPerM":4,"cacheReadPerM":0.11},{"from":"2026-11-04","to":"2026-11-05","inputPerM":0.115,"outputPerM":4.6,"cacheReadPerM":0.11},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.103,"outputPerM":4.11,"cacheReadPerM":0.098},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.1165,"outputPerM":4.66,"cacheReadPerM":0.111}]},{"id":"elorian/glimpse","name":"Glimpse","vendor":"elorian","releasedAt":"2026-10-17","contextLength":262144,"maxOutput":256000,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Glimpse is a multimodal visual reasoning model from Elorian, built for visual understanding and complex visual reasoning in the physical world. It reasons directly over images and video rather than...","inputPerM":0.6,"outputPerM":3,"cacheReadPerM":0.06,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":3,"cacheReadPerM":0.06}]},{"id":"fireworks/ember-1","name":"Ember-1","vendor":"fireworks","releasedAt":"2026-09-24","contextLength":1048576,"maxOutput":943718,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Ember-1 is a specialized reasoning model from Fireworks Research, built on [Kimi K3](https://openrouter.ai/moonshotai/kimi-k3). It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40%...","inputPerM":3,"outputPerM":15,"cacheReadPerM":0.3,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":15,"cacheReadPerM":0.3}]},{"id":"google/gemini-2.5-flash","name":"Gemini 2.5 Flash","vendor":"google","releasedAt":"2025-06-17","contextLength":1048576,"maxOutput":65535,"input":["file","image","text","audio","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in \"thinking\" capabilities, enabling it to provide responses with greater...","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":0.03}]},{"id":"google/gemini-2.5-flash-image","name":"Nano Banana (Gemini 2.5 Flash Image)","vendor":"google","releasedAt":"2025-10-07","contextLength":32768,"maxOutput":8192,"input":["image","text"],"output":["image","text"],"features":["structured_outputs"],"description":"Gemini 2.5 Flash Image, a.k.a. \"Nano Banana,\" is now generally available. It is a state of the art image generation model with contextual understanding. It is capable of image generation,...","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":0.03}]},{"id":"google/gemini-2.5-flash-lite","name":"Gemini 2.5 Flash Lite","vendor":"google","releasedAt":"2025-07-22","contextLength":1048576,"maxOutput":65535,"input":["text","image","file","audio","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance...","inputPerM":0.1,"outputPerM":0.4,"cacheReadPerM":0.01,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.4,"cacheReadPerM":0.01}]},{"id":"google/gemini-2.5-pro","name":"Gemini 2.5 Pro","vendor":"google","releasedAt":"2025-06-17","contextLength":1048576,"maxOutput":65536,"input":["text","image","file","audio","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125}]},{"id":"google/gemini-2.5-pro-preview","name":"Gemini 2.5 Pro Preview 06-05","vendor":"google","releasedAt":"2025-06-05","contextLength":1048576,"maxOutput":65536,"input":["file","image","text","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125}]},{"id":"google/gemini-3.1-flash-image","name":"Nano Banana 2 (Gemini 3.1 Flash Image)","vendor":"google","releasedAt":"2026-06-18","contextLength":131072,"maxOutput":32768,"input":["image","text"],"output":["image","text"],"features":["reasoning","structured_outputs"],"description":"Gemini 3.1 Flash Image, a.k.a. \"Nano Banana 2,\" is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines advanced...","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":null}]},{"id":"google/gemini-3.1-flash-image-preview","name":"Nano Banana 2 (Gemini 3.1 Flash Image Preview)","vendor":"google","releasedAt":"2026-02-26","contextLength":65536,"maxOutput":58982,"input":["image","text"],"output":["image","text"],"features":["reasoning","structured_outputs"],"description":"Gemini 3.1 Flash Image Preview, a.k.a. \"Nano Banana 2,\" is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines...","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":null}]},{"id":"google/gemini-3.1-flash-lite","name":"Gemini 3.1 Flash Lite","vendor":"google","releasedAt":"2026-05-07","contextLength":1048576,"maxOutput":65536,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads. It supports text, image, video, audio, and PDF inputs, and is designed for lightweight agentic...","inputPerM":0.25,"outputPerM":1.5,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":1.5,"cacheReadPerM":0.025}]},{"id":"google/gemini-3.1-flash-lite-image","name":"Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)","vendor":"google","releasedAt":"2026-06-30","contextLength":65536,"maxOutput":58982,"input":["image","text"],"output":["image","text"],"features":["reasoning"],"description":"Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) is Google's fastest, most cost-efficient Gemini image model, built for high-velocity developer pipelines and rapid-fire visual exploration. It delivers text-to-image generation...","inputPerM":0.25,"outputPerM":1.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":1.5,"cacheReadPerM":null}]},{"id":"google/gemini-3.1-flash-lite-preview","name":"Gemini 3.1 Flash Lite Preview","vendor":"google","releasedAt":"2026-03-03","contextLength":1048576,"maxOutput":65536,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases. It outperforms Gemini 2.5 Flash Lite on overall quality and approaches Gemini 2.5 Flash performance across...","inputPerM":0.25,"outputPerM":1.5,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":1.5,"cacheReadPerM":0.025}]},{"id":"google/gemini-3.1-pro-preview","name":"Gemini 3.1 Pro Preview","vendor":"google","releasedAt":"2026-02-19","contextLength":1048576,"maxOutput":65536,"input":["audio","file","image","text","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation...","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2}]},{"id":"google/gemini-3.1-pro-preview-customtools","name":"Gemini 3.1 Pro Preview Custom Tools","vendor":"google","releasedAt":"2026-02-25","contextLength":1048576,"maxOutput":65536,"input":["text","audio","image","video","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool when more efficient third-party...","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2}]},{"id":"google/gemini-3.5-flash","name":"Gemini 3.5 Flash","vendor":"google","releasedAt":"2026-05-19","contextLength":1048576,"maxOutput":65536,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution...","inputPerM":1.5,"outputPerM":9,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.5,"outputPerM":9,"cacheReadPerM":0.15}]},{"id":"google/gemini-3.5-flash-lite","name":"Gemini 3.5 Flash Lite","vendor":"google","releasedAt":"2026-07-21","contextLength":1048576,"maxOutput":65536,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":0.03}]},{"id":"google/gemini-3.6-flash","name":"Gemini 3.6 Flash","vendor":"google","releasedAt":"2026-07-21","contextLength":1048576,"maxOutput":65536,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development. It is designed to produce polished outputs with fewer unnecessary edits and...","inputPerM":0.75,"outputPerM":3.75,"cacheReadPerM":0.075,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.75,"outputPerM":3.75,"cacheReadPerM":0.075}]},{"id":"google/gemini-3.7-flash","name":"Gemini 3.7 Flash","vendor":"google","releasedAt":"2026-08-13","contextLength":1048576,"maxOutput":65536,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...","inputPerM":0.75,"outputPerM":3.75,"cacheReadPerM":0.075,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.75,"outputPerM":3.75,"cacheReadPerM":0.075}]},{"id":"google/gemini-3.8-flash","name":"Gemini 3.8 Flash","vendor":"google","releasedAt":"2026-09-02","contextLength":1048576,"maxOutput":65536,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.","inputPerM":0.825,"outputPerM":4.125,"cacheReadPerM":0.0825,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.825,"outputPerM":4.125,"cacheReadPerM":0.0825}]},{"id":"google/gemini-3-flash-preview","name":"Gemini 3 Flash Preview","vendor":"google","releasedAt":"2025-12-17","contextLength":1048576,"maxOutput":65536,"input":["text","image","file","audio","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance. It delivers near Pro level reasoning and tool...","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":3,"cacheReadPerM":0.05}]},{"id":"google/gemini-3-pro-image","name":"Nano Banana Pro (Gemini 3 Pro Image)","vendor":"google","releasedAt":"2026-06-18","contextLength":131072,"maxOutput":32768,"input":["image","text"],"output":["image","text"],"features":["reasoning","structured_outputs","tools"],"description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro. It extends the original Nano Banana with significantly improved multimodal reasoning, real-world grounding, and...","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2}]},{"id":"google/gemini-3-pro-image-preview","name":"Nano Banana Pro (Gemini 3 Pro Image Preview)","vendor":"google","releasedAt":"2025-11-20","contextLength":65536,"maxOutput":32768,"input":["image","text"],"output":["image","text"],"features":["reasoning","structured_outputs"],"description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro. It extends the original Nano Banana with significantly improved multimodal reasoning, real-world grounding, and...","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2}]},{"id":"google/gemini-4-argon","name":"Gemini 4 Argon","vendor":"google","releasedAt":"2026-10-13","contextLength":1048576,"maxOutput":943718,"input":["audio","file","image","text","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1}]},{"id":"google/gemini-nano-banana-2.1","name":"Nano Banana 2.1","vendor":"google","releasedAt":"2026-10-06","contextLength":65536,"maxOutput":58982,"input":["image","text"],"output":["image","text"],"features":["reasoning","structured_outputs","tools"],"description":"Nano Banana 2.1 (Gemini Nano Banana 2.1) is Google's image generation and editing model on the Flash tier, succeeding Nano Banana 2 and Nano Banana Pro. It improves product recontextualization,...","inputPerM":1.5,"outputPerM":7.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.5,"outputPerM":7.5,"cacheReadPerM":null}]},{"id":"google/gemini-omni-1.1-flash","name":"Gemini Omni 1.1 Flash","vendor":"google","releasedAt":"2026-11-04","contextLength":131072,"maxOutput":65536,"input":["text","image","video"],"output":["text","video"],"features":["reasoning"],"description":"Gemini Omni 1.1 Flash is Google's native multimodal video generation and editing model, the generally available successor to Gemini Omni Flash Preview. It is suited for text-to-video, image-to-video, first and...","inputPerM":1.5,"outputPerM":9,"cacheReadPerM":null,"firstSeen":"2026-11-05","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-11-05","to":"2026-11-06","inputPerM":1.5,"outputPerM":9,"cacheReadPerM":null}]},{"id":"google/gemini-omni-1.2-flash","name":"Gemini Omni 1.2 Flash","vendor":"google","releasedAt":"2026-11-05","contextLength":131072,"maxOutput":65536,"input":["text","image","video"],"output":["text","video"],"features":["reasoning"],"description":"Gemini Omni 1.2 Flash is Google's native multimodal video generation and editing model, the successor to Gemini Omni 1.1 Flash. It generates clips up to 15 seconds long with native...","inputPerM":1.5,"outputPerM":9,"cacheReadPerM":null,"firstSeen":"2026-11-05","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-11-05","to":"2026-11-06","inputPerM":1.5,"outputPerM":9,"cacheReadPerM":null}]},{"id":"google/gemma-2-27b-it","name":"Gemma 2 27B","vendor":"google","releasedAt":"2024-07-13","contextLength":8192,"maxOutput":2048,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini). Gemma models are well-suited for a variety of...","inputPerM":0.65,"outputPerM":0.65,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.65,"outputPerM":0.65,"cacheReadPerM":null}]},{"id":"google/gemma-3-12b-it","name":"Gemma 3 12B","vendor":"google","releasedAt":"2025-03-13","contextLength":131072,"maxOutput":16384,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...","inputPerM":0.05,"outputPerM":0.15,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.15,"cacheReadPerM":null}]},{"id":"google/gemma-3-27b-it","name":"Gemma 3 27B","vendor":"google","releasedAt":"2025-03-12","contextLength":131072,"maxOutput":117964,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...","inputPerM":0.08,"outputPerM":0.35,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.08,"outputPerM":0.35,"cacheReadPerM":0.04}]},{"id":"google/gemma-3-4b-it","name":"Gemma 3 4B","vendor":"google","releasedAt":"2025-03-13","contextLength":131072,"maxOutput":16384,"input":["text","image"],"output":["text"],"features":["structured_outputs"],"description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...","inputPerM":0.05,"outputPerM":0.1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.1,"cacheReadPerM":null}]},{"id":"google/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B","vendor":"google","releasedAt":"2026-04-03","contextLength":262144,"maxOutput":235929,"input":["image","text","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...","inputPerM":0.09,"outputPerM":0.3,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.09,"outputPerM":0.3,"cacheReadPerM":0.05},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.0765,"outputPerM":0.255,"cacheReadPerM":0.0425},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.09,"outputPerM":0.3,"cacheReadPerM":0.05},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.0765,"outputPerM":0.255,"cacheReadPerM":0.0425},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.09,"outputPerM":0.3,"cacheReadPerM":0.05},{"from":"2026-10-31","to":"2026-11-01","inputPerM":0.0675,"outputPerM":0.225,"cacheReadPerM":0.0375},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.023,"outputPerM":0.07,"cacheReadPerM":0.012},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.058,"outputPerM":0.173,"cacheReadPerM":0.029},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.068,"outputPerM":0.205,"cacheReadPerM":0.034},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.0765,"outputPerM":0.255,"cacheReadPerM":0.0425},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.039,"outputPerM":0.117,"cacheReadPerM":0.019},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.082,"outputPerM":0.246,"cacheReadPerM":0.041},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.086,"outputPerM":0.257,"cacheReadPerM":0.043},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.07,"outputPerM":0.211,"cacheReadPerM":0.035},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.078,"outputPerM":0.233,"cacheReadPerM":0.039},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":0.05},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.027,"outputPerM":0.08,"cacheReadPerM":0.013},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.027,"outputPerM":0.081,"cacheReadPerM":0.014},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.047,"outputPerM":0.142,"cacheReadPerM":0.024},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.013,"outputPerM":0.04,"cacheReadPerM":0.007},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.0765,"outputPerM":0.255,"cacheReadPerM":0.0425},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.09,"outputPerM":0.3,"cacheReadPerM":0.05}]},{"id":"google/gemma-4-31b-it","name":"Gemma 4 31B","vendor":"google","releasedAt":"2026-04-02","contextLength":262144,"maxOutput":16384,"input":["image","text","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...","inputPerM":0.2,"outputPerM":0.4,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":0.4,"cacheReadPerM":null}]},{"id":"google/lyria-3-clip-preview","name":"Lyria 3 Clip Preview","vendor":"google","releasedAt":"2026-03-30","contextLength":1048576,"maxOutput":65536,"input":["text","image"],"output":["text","audio"],"features":[],"description":"30 second duration clips are priced at $0.04 per clip. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate...","inputPerM":0,"outputPerM":0,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0,"outputPerM":0,"cacheReadPerM":null}]},{"id":"google/lyria-3-pro-preview","name":"Lyria 3 Pro Preview","vendor":"google","releasedAt":"2026-03-30","contextLength":1048576,"maxOutput":65536,"input":["text","image"],"output":["text","audio"],"features":[],"description":"Full-length songs are priced at $0.08 per song. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate high-quality, 48kHz...","inputPerM":0,"outputPerM":0,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0,"outputPerM":0,"cacheReadPerM":null}]},{"id":"gryphe/mythomax-l2-13b","name":"MythoMax 13B","vendor":"gryphe","releasedAt":"2023-07-02","contextLength":8192,"maxOutput":3686,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge","inputPerM":0.08,"outputPerM":0.11,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.08,"outputPerM":0.11,"cacheReadPerM":null}]},{"id":"ibm-granite/granite-4.2-8b","name":"Granite 4.2 8B","vendor":"ibm-granite","releasedAt":"2026-08-31","contextLength":131072,"maxOutput":117964,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Granite 4.2 8B is a dense reasoning model from IBM. It is suited for mathematics, code generation, multilingual dialogue, and agentic workflows that need multi-step reasoning. It supports full, low-effort,...","inputPerM":0.06,"outputPerM":0.25,"cacheReadPerM":0.015,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.06,"outputPerM":0.25,"cacheReadPerM":0.015},{"from":"2026-10-29","to":"2026-11-03","inputPerM":0.06,"outputPerM":0.15,"cacheReadPerM":0.015},{"from":"2026-11-04","to":"2026-11-06","inputPerM":0.06,"outputPerM":0.25,"cacheReadPerM":0.015}]},{"id":"inception/mercury-2","name":"Mercury 2","vendor":"inception","releasedAt":"2026-03-04","contextLength":128000,"maxOutput":50000,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM). Instead of generating tokens sequentially, Mercury 2 produces and refines multiple tokens in parallel, achieving...","inputPerM":0.25,"outputPerM":0.75,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":0.75,"cacheReadPerM":0.025}]},{"id":"inception/mercury-2.5","name":"Mercury 2.5","vendor":"inception","releasedAt":"2026-09-08","contextLength":260000,"maxOutput":65536,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception. Instead of generating tokens sequentially, Mercury 2.5 produces and refines multiple tokens in parallel, achieving...","inputPerM":0.04,"outputPerM":0.15,"cacheReadPerM":0.004,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.04,"outputPerM":0.15,"cacheReadPerM":0.004}]},{"id":"inception/mercury-3","name":"Mercury 3","vendor":"inception","releasedAt":"2026-10-16","contextLength":1000000,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Mercury 3 is a diffusion-based language model from Inception, and the lab's most capable model to date. It is suited for agentic workflows and real-time applications where latency compounds, such...","inputPerM":0.1,"outputPerM":0.35,"cacheReadPerM":0.01,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.35,"cacheReadPerM":0.01}]},{"id":"inclusionai/ling-3.0-flash","name":"Ling 3.0 Flash","vendor":"inclusionai","releasedAt":"2026-07-23","contextLength":262144,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"*Ling-3.0-flash* is a *124B-parameter Mixture-of-Experts (MoE) model*, with approximately *5.1B parameters activated per token*. The model is designed with *token efficiency and production-scale agentic inference* as key priorities, enabling developers...","inputPerM":0.021,"outputPerM":0.063,"cacheReadPerM":0.0042,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.021,"outputPerM":0.063,"cacheReadPerM":0.0042}]},{"id":"inclusionai/ling-3.0-flash-vl","name":"Ling 3.0 Flash VL","vendor":"inclusionai","releasedAt":"2026-09-10","contextLength":131072,"maxOutput":32768,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities while adding native visual perception and advanced visual...","inputPerM":0.06,"outputPerM":0.18,"cacheReadPerM":0.012,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.06,"outputPerM":0.18,"cacheReadPerM":0.012}]},{"id":"inclusionai/ling-3.0-tiny","name":"Ling 3.0 Tiny","vendor":"inclusionai","releasedAt":"2026-08-06","contextLength":262144,"maxOutput":250000,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Ling 3.0 Tiny is a mixture-of-experts model from InclusionAI, with 1.3B active parameters out of 7.9B total. It is designed for responsive agents, instruction following, and multi-turn conversations, with switchable...","inputPerM":0.015,"outputPerM":0.045,"cacheReadPerM":0.003,"firstSeen":"2026-11-05","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.015,"outputPerM":0.045,"cacheReadPerM":0.003}]},{"id":"inclusionai/ling-3.1-flash","name":"Ling 3.1 Flash","vendor":"inclusionai","releasedAt":"2026-10-02","contextLength":1000000,"maxOutput":262144,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Ling 3.1 Flash is a hybrid reasoning mixture-of-experts model from inclusionAI, with 25B active parameters out of 560B total.","inputPerM":0.075,"outputPerM":0.225,"cacheReadPerM":0.015,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.075,"outputPerM":0.225,"cacheReadPerM":0.015}]},{"id":"inference-net/schematron-v2-small","name":"Schematron V2 Small","vendor":"inference-net","releasedAt":"2026-09-12","contextLength":128000,"maxOutput":8192,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Schematron V2 Small is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes extraction quality for complex schemas and long pages. Extraction instructions must be supplied through a JSON schema...","inputPerM":0.05,"outputPerM":0.23,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.23,"cacheReadPerM":0.05}]},{"id":"inference-net/schematron-v2-turbo","name":"Schematron V2 Turbo","vendor":"inference-net","releasedAt":"2026-09-12","contextLength":128000,"maxOutput":8192,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Schematron V2 Turbo is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes throughput for high-volume extraction workloads. Extraction instructions must be supplied through a JSON schema in response_format rather...","inputPerM":0.03,"outputPerM":0.15,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.03,"outputPerM":0.15,"cacheReadPerM":0.03}]},{"id":"liquid/lfm-2.5-24b-a2b","name":"LFM2.5-24B-A2B","vendor":"liquid","releasedAt":"2026-10-13","contextLength":128000,"maxOutput":8192,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"LFM2.5-24B-A2B is a mixture-of-experts reasoning model from Liquid AI with 24B total and 2.45B active parameters. It always reasons before answering and is built for agents that call tools. Liquid...","inputPerM":0.05,"outputPerM":0.7,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.7,"cacheReadPerM":0.025}]},{"id":"liquid/lfm-2.5-2.6b","name":"LFM2.5-2.6B","vendor":"liquid","releasedAt":"2026-08-11","contextLength":128000,"maxOutput":8192,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"LFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or...","inputPerM":0.35,"outputPerM":0.5,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.35,"outputPerM":0.5,"cacheReadPerM":0.04}]},{"id":"mancer/weaver","name":"Weaver (alpha)","vendor":"mancer","releasedAt":"2023-08-02","contextLength":8000,"maxOutput":6000,"input":["text"],"output":["text"],"features":[],"description":"An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory. Meant for use in roleplay/narrative situations.","inputPerM":0.4,"outputPerM":0.75,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.4,"outputPerM":0.75,"cacheReadPerM":null}]},{"id":"meituan/longcat-2.0","name":"LongCat 2.0","vendor":"meituan","releasedAt":"2026-07-20","contextLength":1048756,"maxOutput":262144,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"LongCat 2.0 is a sparse mixture-of-experts language model from Meituan, with 48B active parameters out of 1.6T total. It is suited for coding, repository-level changes, long-horizon problem solving, and agentic...","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.006}]},{"id":"meta/muse-glimmer-30b","name":"Muse Glimmer 30B","vendor":"meta","releasedAt":"2026-08-09","contextLength":131072,"maxOutput":16384,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.04}]},{"id":"meta/muse-spark-1.1","name":"Muse Spark 1.1","vendor":"meta","releasedAt":"2026-07-16","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. It accepts text, images, video, and PDF documents and returns text, with a 1M-token context window....","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15}]},{"id":"meta/muse-spark-1.2","name":"Muse Spark 1.2","vendor":"meta","releasedAt":"2026-08-05","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Spark 1.2 is a reasoning model from Meta, designed for complex agentic tasks. It accepts text, images, video, and PDF documents, returns text, and offers a 1M-token context window....","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15}]},{"id":"meta/muse-spark-1.2-contributor","name":"Muse Spark 1.2 Contributor","vendor":"meta","releasedAt":"2026-08-21","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Spark 1.2 contributor tier is a reasoning model from Meta designed for developers who want to start building at an even lower cost. It’s meaningfully cheaper than Muse Spark...","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":0.002,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":0.002}]},{"id":"meta/muse-spark-1.3","name":"Muse Spark 1.3","vendor":"meta","releasedAt":"2026-09-02","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It is designed to keep track of information across extended tasks, work through...","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15}]},{"id":"meta/muse-spark-1.3-contributor","name":"Muse Spark 1.3 Contributor","vendor":"meta","releasedAt":"2026-09-02","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Spark 1.3 Contributor is the cost-efficient contributor tier of Meta’s multimodal reasoning model for experimentation, learning, and early-stage agentic, multi-agent, and coding workflows. It is designed to track information...","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":0.002,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":0.002}]},{"id":"meta/muse-spark-open","name":"Muse Spark Open","vendor":"meta","releasedAt":"2026-10-15","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":4.25,"cacheReadPerM":0.15}]},{"id":"meta/muse-wonder","name":"Muse Wonder","vendor":"meta","releasedAt":"2026-10-12","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with...","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.2}]},{"id":"meta-llama/llama-3.1-70b-instruct","name":"Llama 3.1 70B Instruct","vendor":"meta-llama","releasedAt":"2024-07-23","contextLength":131072,"maxOutput":8192,"input":["text"],"output":["text"],"features":[],"description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 70B instruct-tuned version is optimized for high quality dialogue usecases. It has demonstrated strong...","inputPerM":0.72,"outputPerM":0.72,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-03","inputPerM":0.4,"outputPerM":0.4,"cacheReadPerM":null},{"from":"2026-11-03","to":"2026-11-06","inputPerM":0.72,"outputPerM":0.72,"cacheReadPerM":null}]},{"id":"meta-llama/llama-3.1-8b-instruct","name":"Llama 3.1 8B Instruct","vendor":"meta-llama","releasedAt":"2024-07-23","contextLength":131072,"maxOutput":117964,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient. It has demonstrated strong performance compared to...","inputPerM":0.05,"outputPerM":0.08,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.02,"outputPerM":0.04,"cacheReadPerM":null},{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.08,"cacheReadPerM":0.025}]},{"id":"meta-llama/llama-3.2-1b-instruct","name":"Llama 3.2 1B Instruct","vendor":"meta-llama","releasedAt":"2024-09-25","contextLength":60000,"maxOutput":54000,"input":["text"],"output":["text"],"features":[],"description":"Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis. Its smaller size allows it to operate...","inputPerM":0.027,"outputPerM":0.201,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.027,"outputPerM":0.201,"cacheReadPerM":null}]},{"id":"meta-llama/llama-3.2-3b-instruct","name":"Llama 3.2 3B Instruct","vendor":"meta-llama","releasedAt":"2024-09-25","contextLength":131072,"maxOutput":72000,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it...","inputPerM":0.0509,"outputPerM":0.335,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-01","inputPerM":0.05,"outputPerM":0.33,"cacheReadPerM":null},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.0509,"outputPerM":0.335,"cacheReadPerM":null},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.05,"outputPerM":0.33,"cacheReadPerM":null},{"from":"2026-11-03","to":"2026-11-06","inputPerM":0.0509,"outputPerM":0.335,"cacheReadPerM":null}]},{"id":"meta-llama/llama-3.3-70b-instruct","name":"Llama 3.3 70B Instruct","vendor":"meta-llama","releasedAt":"2024-12-06","contextLength":131072,"maxOutput":32768,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...","inputPerM":0.59,"outputPerM":0.79,"cacheReadPerM":0.295,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-02","inputPerM":0.1,"outputPerM":0.32,"cacheReadPerM":null},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.2,"outputPerM":0.52,"cacheReadPerM":0.1},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.1,"outputPerM":0.32,"cacheReadPerM":null},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.2,"outputPerM":0.52,"cacheReadPerM":0.1},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.59,"outputPerM":0.79,"cacheReadPerM":0.295}]},{"id":"meta-llama/llama-4-maverick","name":"Llama 4 Maverick","vendor":"meta-llama","releasedAt":"2025-04-05","contextLength":524288,"maxOutput":32768,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...","inputPerM":0.35,"outputPerM":1,"cacheReadPerM":0.17,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-04","inputPerM":0.1875,"outputPerM":0.6525,"cacheReadPerM":0.05},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.35,"outputPerM":1,"cacheReadPerM":0.17}]},{"id":"meta-llama/llama-4-scout","name":"Llama 4 Scout","vendor":"meta-llama","releasedAt":"2025-04-05","contextLength":1310720,"maxOutput":16384,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":null}]},{"id":"meta-llama/llama-guard-4-12b","name":"Llama Guard 4 12B","vendor":"meta-llama","releasedAt":"2025-04-30","contextLength":163840,"maxOutput":16384,"input":["image","text"],"output":["text"],"features":[],"description":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM...","inputPerM":0.18,"outputPerM":0.18,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.18,"outputPerM":0.18,"cacheReadPerM":null}]},{"id":"microsoft/phi-4","name":"Phi 4","vendor":"microsoft","releasedAt":"2025-01-10","contextLength":16384,"maxOutput":14745,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion...","inputPerM":0.07,"outputPerM":0.14,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.07,"outputPerM":0.14,"cacheReadPerM":null}]},{"id":"minimax/minimax-01","name":"MiniMax-01","vendor":"minimax","releasedAt":"2025-01-15","contextLength":1000192,"maxOutput":900172,"input":["text","image"],"output":["text"],"features":[],"description":"MiniMax-01 is a combines MiniMax-Text-01 for text generation and MiniMax-VL-01 for image understanding. It has 456 billion parameters, with 45.9 billion parameters activated per inference, and can handle a context...","inputPerM":0.2,"outputPerM":1.1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-05","gone":"2026-11-05","history":[{"from":"2026-10-28","to":"2026-11-05","inputPerM":0.2,"outputPerM":1.1,"cacheReadPerM":null}]},{"id":"minimax/minimax-m1","name":"MiniMax M1","vendor":"minimax","releasedAt":"2025-06-17","contextLength":1000000,"maxOutput":40000,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference. It leverages a hybrid Mixture-of-Experts (MoE) architecture paired with a custom \"lightning attention\" mechanism, allowing it...","inputPerM":0.55,"outputPerM":2.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.55,"outputPerM":2.2,"cacheReadPerM":null}]},{"id":"minimax/minimax-m2","name":"MiniMax M2","vendor":"minimax","releasedAt":"2025-10-23","contextLength":204800,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion total), it delivers near-frontier intelligence across general reasoning,...","inputPerM":0.255,"outputPerM":1.02,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-05","gone":"2026-11-05","history":[{"from":"2026-10-28","to":"2026-11-05","inputPerM":0.255,"outputPerM":1.02,"cacheReadPerM":null}]},{"id":"minimax/minimax-m2.1","name":"MiniMax M2.1","vendor":"minimax","releasedAt":"2025-12-23","contextLength":204800,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development. With only 10 billion activated parameters, it delivers a major jump in real-world...","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-05","gone":"2026-11-05","history":[{"from":"2026-10-28","to":"2026-11-05","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.03}]},{"id":"minimax/minimax-m2.5","name":"MiniMax M2.5","vendor":"minimax","releasedAt":"2026-02-12","contextLength":204800,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1...","inputPerM":0.27,"outputPerM":0.95,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-30","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.03},{"from":"2026-10-31","to":"2026-11-06","inputPerM":0.27,"outputPerM":0.95,"cacheReadPerM":0.03}]},{"id":"minimax/minimax-m2.7","name":"MiniMax M2.7","vendor":"minimax","releasedAt":"2026-03-18","contextLength":204800,"maxOutput":176947,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement. Built to actively participate in its own evolution, M2.7 integrates advanced agentic capabilities through multi-agent...","inputPerM":0.21,"outputPerM":0.84,"cacheReadPerM":0.042,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-05","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.06},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.21,"outputPerM":0.84,"cacheReadPerM":0.042}]},{"id":"minimax/minimax-m2-her","name":"MiniMax M2-her","vendor":"minimax","releasedAt":"2026-01-23","contextLength":65536,"maxOutput":2048,"input":["text"],"output":["text"],"features":[],"description":"MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations. Designed to stay consistent in tone and personality, it supports rich message...","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-05","gone":"2026-11-05","history":[{"from":"2026-10-28","to":"2026-11-05","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.03}]},{"id":"minimax/minimax-m3","name":"MiniMax M3","vendor":"minimax","releasedAt":"2026-05-31","contextLength":1048576,"maxOutput":512000,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.06,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":1.2,"cacheReadPerM":0.06}]},{"id":"minimax/minimax-m3.1-flash","name":"MiniMax M3.1 Flash","vendor":"minimax","releasedAt":"2026-10-19","contextLength":1048576,"maxOutput":524288,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it delivers significant improvements across multiple benchmarks, particularly in complex agentic tasks, coding, and cybersecurity. It is built for complex,...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-30","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015},{"from":"2026-10-30","to":"2026-10-31","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0075},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0075},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0075},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0075},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0075},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0075},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0075},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015}]},{"id":"mistralai/codestral-2508","name":"Codestral 2508","vendor":"mistralai","releasedAt":"2025-08-01","contextLength":256000,"maxOutput":204800,"input":["text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral's cutting-edge language model for coding released end of July 2025. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.\n\n[Blog Post](https://mistral.ai/news/codestral-25-08)","inputPerM":0.3,"outputPerM":0.9,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":0.9,"cacheReadPerM":0.03}]},{"id":"mistralai/devstral-2512","name":"Devstral 2 2512","vendor":"mistralai","releasedAt":"2025-12-09","contextLength":262144,"maxOutput":209715,"input":["text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring...","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.04}]},{"id":"mistralai/ministral-14b-2512","name":"Ministral 3 14B 2512","vendor":"mistralai","releasedAt":"2025-12-02","contextLength":262144,"maxOutput":209715,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart. A powerful and efficient language...","inputPerM":0.2,"outputPerM":0.2,"cacheReadPerM":0.02,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":0.2,"cacheReadPerM":0.02}]},{"id":"mistralai/ministral-3b-2512","name":"Ministral 3 3B 2512","vendor":"mistralai","releasedAt":"2025-12-02","contextLength":131072,"maxOutput":104857,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","inputPerM":0.1,"outputPerM":0.1,"cacheReadPerM":0.01,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.1,"cacheReadPerM":0.01}]},{"id":"mistralai/ministral-8b-2512","name":"Ministral 3 8B 2512","vendor":"mistralai","releasedAt":"2025-12-02","contextLength":262144,"maxOutput":209715,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","inputPerM":0.15,"outputPerM":0.15,"cacheReadPerM":0.015,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.15,"cacheReadPerM":0.015}]},{"id":"mistralai/mistral-large","name":"Mistral Large","vendor":"mistralai","releasedAt":"2024-02-26","contextLength":128000,"maxOutput":102400,"input":["text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/)....","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.2}]},{"id":"mistralai/mistral-large-2407","name":"Mistral Large 2407","vendor":"mistralai","releasedAt":"2024-11-19","contextLength":131072,"maxOutput":104857,"input":["text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/)....","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.2}]},{"id":"mistralai/mistral-large-2512","name":"Mistral Large 3 2512","vendor":"mistralai","releasedAt":"2025-12-01","contextLength":262144,"maxOutput":209715,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","inputPerM":0.5,"outputPerM":1.5,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":1.5,"cacheReadPerM":0.05}]},{"id":"mistralai/mistral-large-4-0","name":"Mistral Large 4","vendor":"mistralai","releasedAt":"2026-10-06","contextLength":1048576,"maxOutput":262144,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads. It offers a 1M-token context window with up...","inputPerM":0.7,"outputPerM":2.1,"cacheReadPerM":0.07,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-29","inputPerM":1.36,"outputPerM":4.18,"cacheReadPerM":0.14},{"from":"2026-10-29","to":"2026-11-06","inputPerM":0.7,"outputPerM":2.1,"cacheReadPerM":0.07}]},{"id":"mistralai/mistral-medium-3","name":"Mistral Medium 3","vendor":"mistralai","releasedAt":"2025-05-07","contextLength":131072,"maxOutput":104857,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost...","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.04}]},{"id":"mistralai/mistral-medium-3.1","name":"Mistral Medium 3.1","vendor":"mistralai","releasedAt":"2025-08-13","contextLength":131072,"maxOutput":104857,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances...","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.04}]},{"id":"mistralai/mistral-medium-3-5","name":"Mistral Medium 3.5","vendor":"mistralai","releasedAt":"2026-04-30","contextLength":262144,"maxOutput":209715,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...","inputPerM":1.5,"outputPerM":7.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.5,"outputPerM":7.5,"cacheReadPerM":null}]},{"id":"mistralai/mistral-nemo","name":"Mistral Nemo","vendor":"mistralai","releasedAt":"2024-07-19","contextLength":131072,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese,...","inputPerM":0.029,"outputPerM":0.03,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.029,"outputPerM":0.03,"cacheReadPerM":null}]},{"id":"mistralai/mistral-saba","name":"Saba","vendor":"mistralai","releasedAt":"2025-02-17","contextLength":32768,"maxOutput":26214,"input":["text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. Trained on curated regional...","inputPerM":0.2,"outputPerM":0.6,"cacheReadPerM":0.02,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":0.6,"cacheReadPerM":0.02}]},{"id":"mistralai/mistral-small-24b-instruct-2501","name":"Mistral Small 3","vendor":"mistralai","releasedAt":"2025-01-30","contextLength":32768,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed...","inputPerM":0.05,"outputPerM":0.08,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.08,"cacheReadPerM":null}]},{"id":"mistralai/mistral-small-2603","name":"Mistral Small 4","vendor":"mistralai","releasedAt":"2026-03-16","contextLength":262144,"maxOutput":209715,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system. It combines strong reasoning from...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015}]},{"id":"mistralai/mistral-small-3.1-24b-instruct","name":"Mistral Small 3.1 24B","vendor":"mistralai","releasedAt":"2025-03-17","contextLength":128000,"maxOutput":102400,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities. It provides state-of-the-art performance in text-based reasoning and...","inputPerM":0.351,"outputPerM":0.555,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.351,"outputPerM":0.555,"cacheReadPerM":null}]},{"id":"mistralai/mistral-small-3.2-24b-instruct","name":"Mistral Small 3.2 24B","vendor":"mistralai","releasedAt":"2025-06-20","contextLength":256000,"maxOutput":16384,"input":["image","text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling. Compared to the 3.1 release, version 3.2 significantly improves accuracy on...","inputPerM":0.075,"outputPerM":0.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.075,"outputPerM":0.2,"cacheReadPerM":null}]},{"id":"mistralai/mixtral-8x22b-instruct","name":"Mixtral 8x22B Instruct","vendor":"mistralai","releasedAt":"2024-04-17","contextLength":65536,"maxOutput":52428,"input":["text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include: - strong math, coding,...","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.2}]},{"id":"mistralai/voxtral-small-24b-2507","name":"Voxtral Small 24B 2507","vendor":"mistralai","releasedAt":"2025-10-30","contextLength":32768,"maxOutput":26214,"input":["text","audio","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio...","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":0.01,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":0.01}]},{"id":"moondream/moondream3.1-9b-a2b","name":"Moondream 3.1","vendor":"moondream","releasedAt":"2026-10-15","contextLength":32000,"maxOutput":4096,"input":["text","image"],"output":["text"],"features":["reasoning"],"description":"Moondream 3.1 is a vision-language model from Moondream, the successor to Moondream 3 Preview, with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and...","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":null}]},{"id":"moondream/moondream3-preview","name":"Moondream 3 Preview","vendor":"moondream","releasedAt":"2026-10-15","contextLength":32000,"maxOutput":4096,"input":["text","image"],"output":["text"],"features":["reasoning"],"description":"Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over images, with...","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":2.5,"cacheReadPerM":null}]},{"id":"moonshotai/kimi-k2","name":"Kimi K2 0711","vendor":"moonshotai","releasedAt":"2025-07-11","contextLength":131072,"maxOutput":98304,"input":["text"],"output":["text"],"features":["tools"],"description":"Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...","inputPerM":0.57,"outputPerM":2.3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.57,"outputPerM":2.3,"cacheReadPerM":null}]},{"id":"moonshotai/kimi-k2-0905","name":"Kimi K2 0905","vendor":"moonshotai","releasedAt":"2025-09-04","contextLength":262144,"maxOutput":98304,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2). It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32...","inputPerM":0.6,"outputPerM":2.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":2.5,"cacheReadPerM":null}]},{"id":"moonshotai/kimi-k2.5","name":"Kimi K2.5","vendor":"moonshotai","releasedAt":"2026-01-27","contextLength":262144,"maxOutput":235929,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...","inputPerM":0.5,"outputPerM":2.5,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":2.5,"cacheReadPerM":0.15}]},{"id":"moonshotai/kimi-k2.6","name":"Kimi K2.6","vendor":"moonshotai","releasedAt":"2026-04-20","contextLength":262144,"maxOutput":235929,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and...","inputPerM":0.65,"outputPerM":3.41,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.65,"outputPerM":3.41,"cacheReadPerM":0.15}]},{"id":"moonshotai/kimi-k2.7-code","name":"Kimi K2.7 Code","vendor":"moonshotai","releasedAt":"2026-06-12","contextLength":262144,"maxOutput":235929,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts...","inputPerM":0.6,"outputPerM":3.5,"cacheReadPerM":0.16,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-30","inputPerM":0.56,"outputPerM":3.1,"cacheReadPerM":0.15},{"from":"2026-10-30","to":"2026-11-05","inputPerM":0.6,"outputPerM":3.5,"cacheReadPerM":0.15},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.6,"outputPerM":3.5,"cacheReadPerM":0.16}]},{"id":"moonshotai/kimi-k2-thinking","name":"Kimi K2 Thinking","vendor":"moonshotai","releasedAt":"2025-11-06","contextLength":262144,"maxOutput":98304,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in...","inputPerM":0.6,"outputPerM":2.5,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":2.5,"cacheReadPerM":0.15}]},{"id":"moonshotai/kimi-k3","name":"Kimi K3","vendor":"moonshotai","releasedAt":"2026-07-16","contextLength":1048576,"maxOutput":943718,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...","inputPerM":0.192,"outputPerM":16,"cacheReadPerM":0.182,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.189,"outputPerM":20,"cacheReadPerM":0.179},{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.201,"outputPerM":20,"cacheReadPerM":0.191},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.192,"outputPerM":20,"cacheReadPerM":0.182},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.192,"outputPerM":15,"cacheReadPerM":0.182},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.239,"outputPerM":10.8,"cacheReadPerM":0.21},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.192,"outputPerM":15,"cacheReadPerM":0.182},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.2,"outputPerM":15,"cacheReadPerM":0.19},{"from":"2026-10-30","to":"2026-11-03","inputPerM":0.192,"outputPerM":15,"cacheReadPerM":0.182},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.276,"outputPerM":9.6,"cacheReadPerM":0.13},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.2,"outputPerM":16,"cacheReadPerM":0.19},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.192,"outputPerM":16,"cacheReadPerM":0.182}]},{"id":"moonshotai/kimi-k3.1-pro","name":"Kimi K3.1 Pro","vendor":"moonshotai","releasedAt":"2026-10-23","contextLength":1048576,"maxOutput":943718,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2}]},{"id":"morph/morph-v3-fast","name":"Morph V3 Fast","vendor":"morph","releasedAt":"2025-07-07","contextLength":262144,"maxOutput":131072,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Morph's fastest apply model for code edits. ~10,500 tokens/sec with 96% accuracy for rapid code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code> <update>{edit_snippet}</update>...","inputPerM":0.8,"outputPerM":1.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.8,"outputPerM":1.2,"cacheReadPerM":null}]},{"id":"morph/morph-v3-large","name":"Morph V3 Large","vendor":"morph","releasedAt":"2025-07-07","contextLength":262144,"maxOutput":131072,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Morph's high-accuracy apply model for complex code edits. ~4,500 tokens/sec with 98% accuracy for precise code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code>...","inputPerM":0.9,"outputPerM":1.9,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.9,"outputPerM":1.9,"cacheReadPerM":null}]},{"id":"nex-agi/nex-n2.5-mini","name":"Nex-N2.5-Mini","vendor":"nex-agi","releasedAt":"2026-09-08","contextLength":262144,"maxOutput":235929,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs"],"description":"Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...","inputPerM":0.025,"outputPerM":0.1,"cacheReadPerM":0.0025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.025,"outputPerM":0.1,"cacheReadPerM":0.0025}]},{"id":"nex-agi/nex-n2.5-pro","name":"Nex-N2.5-Pro","vendor":"nex-agi","releasedAt":"2026-09-08","contextLength":262144,"maxOutput":235929,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...","inputPerM":0.075,"outputPerM":0.25,"cacheReadPerM":0.015,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.075,"outputPerM":0.25,"cacheReadPerM":0.015}]},{"id":"nousresearch/hermes-3-llama-3.1-405b","name":"Hermes 3 405B Instruct","vendor":"nousresearch","releasedAt":"2024-08-16","contextLength":131072,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the...","inputPerM":1,"outputPerM":1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-02","gone":"2026-11-02","history":[{"from":"2026-10-28","to":"2026-11-02","inputPerM":1,"outputPerM":1,"cacheReadPerM":null}]},{"id":"nousresearch/hermes-3-llama-3.1-70b","name":"Hermes 3 70B Instruct","vendor":"nousresearch","releasedAt":"2024-08-18","contextLength":131072,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the...","inputPerM":0.7,"outputPerM":0.7,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-02","gone":"2026-11-02","history":[{"from":"2026-10-28","to":"2026-11-02","inputPerM":0.7,"outputPerM":0.7,"cacheReadPerM":null}]},{"id":"nousresearch/hermes-4-405b","name":"Hermes 4 405B","vendor":"nousresearch","releasedAt":"2025-08-26","contextLength":131072,"maxOutput":117964,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs"],"description":"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with...","inputPerM":1,"outputPerM":3,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":3,"cacheReadPerM":0.2}]},{"id":"nvidia/nemotron-3.5-content-safety","name":"Nemotron 3.5 Content Safety","vendor":"nvidia","releasedAt":"2026-06-04","contextLength":131072,"maxOutput":117964,"input":["text","image"],"output":["text"],"features":["reasoning"],"description":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It moderates both inputs to and responses from LLMs and VLMs, accepting...","inputPerM":0.2,"outputPerM":0.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":0.2,"cacheReadPerM":null}]},{"id":"nvidia/nemotron-3.5-lightning","name":"Nemotron 3.5 Lightning","vendor":"nvidia","releasedAt":"2026-08-11","contextLength":262144,"maxOutput":235929,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...","inputPerM":0.07,"outputPerM":0.2,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.07,"outputPerM":0.2,"cacheReadPerM":0.04}]},{"id":"nvidia/nemotron-3.5-super-vl","name":"Nemotron 3.5 Super VL","vendor":"nvidia","releasedAt":"2026-10-15","contextLength":1048576,"maxOutput":943718,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...","inputPerM":0.15,"outputPerM":0.5,"cacheReadPerM":0.02,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.5,"cacheReadPerM":0.02}]},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"Nemotron 3 Nano 30B A3B","vendor":"nvidia","releasedAt":"2025-12-14","contextLength":262144,"maxOutput":235929,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...","inputPerM":0.06,"outputPerM":0.24,"cacheReadPerM":0.012,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.06,"outputPerM":0.24,"cacheReadPerM":0.012}]},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"Nemotron 3 Super","vendor":"nvidia","releasedAt":"2026-03-11","contextLength":1048576,"maxOutput":262144,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...","inputPerM":0.3,"outputPerM":2.4,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":2.4,"cacheReadPerM":0.15}]},{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"Nemotron 3 Ultra","vendor":"nvidia","releasedAt":"2026-06-04","contextLength":262144,"maxOutput":182520,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...","inputPerM":0.6,"outputPerM":2.4,"cacheReadPerM":0.12,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":2.4,"cacheReadPerM":0.12}]},{"id":"nvidia/switchyard","name":"Switchyard","vendor":"nvidia","releasedAt":"2026-09-21","contextLength":1000000,"maxOutput":null,"input":["text"],"output":["text"],"features":[],"description":"Switchyard is an open-source model routing system that switches between multiple models to optimize the cost of requests. By default it will use OpenRouter market data to select the most...","inputPerM":null,"outputPerM":null,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":null,"outputPerM":null,"cacheReadPerM":null}]},{"id":"openai/gpt-3.5-turbo-0613","name":"GPT-3.5 Turbo (older v0613)","vendor":"openai","releasedAt":"2024-01-25","contextLength":4095,"maxOutput":3685,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks.\n\nTraining data up to Sep 2021.","inputPerM":1,"outputPerM":2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":2,"cacheReadPerM":null}]},{"id":"openai/gpt-3.5-turbo-16k","name":"GPT-3.5 Turbo 16k","vendor":"openai","releasedAt":"2023-08-28","contextLength":16385,"maxOutput":4096,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost. Training data: up...","inputPerM":3,"outputPerM":4,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":4,"cacheReadPerM":null}]},{"id":"openai/gpt-4","name":"GPT-4","vendor":"openai","releasedAt":"2023-05-28","contextLength":8191,"maxOutput":4096,"input":["text"],"output":["text"],"features":["tools"],"description":"OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and advanced reasoning...","inputPerM":30,"outputPerM":60,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":30,"outputPerM":60,"cacheReadPerM":null}]},{"id":"openai/gpt-4.1","name":"GPT-4.1","vendor":"openai","releasedAt":"2025-04-14","contextLength":1047576,"maxOutput":32768,"input":["image","text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning. It supports a 1 million token context window and outperforms GPT-4o and...","inputPerM":2,"outputPerM":8,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":8,"cacheReadPerM":0.5}]},{"id":"openai/gpt-4.1-mini","name":"GPT-4.1 Mini","vendor":"openai","releasedAt":"2025-04-14","contextLength":1047576,"maxOutput":32768,"input":["image","text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost. It retains a 1 million token context window and scores 45.1% on hard...","inputPerM":0.4,"outputPerM":1.6,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.4,"outputPerM":1.6,"cacheReadPerM":0.1}]},{"id":"openai/gpt-4.1-nano","name":"GPT-4.1 Nano","vendor":"openai","releasedAt":"2025-04-14","contextLength":1047576,"maxOutput":942818,"input":["image","text","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million...","inputPerM":0.1,"outputPerM":0.4,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.4,"cacheReadPerM":0.03}]},{"id":"openai/gpt-4o","name":"GPT-4o","vendor":"openai","releasedAt":"2024-05-13","contextLength":128000,"maxOutput":16384,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools","web_search_options"],"description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as...","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":1.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":1.25}]},{"id":"openai/gpt-4o-2024-05-13","name":"GPT-4o (2024-05-13)","vendor":"openai","releasedAt":"2024-05-13","contextLength":128000,"maxOutput":4096,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools","web_search_options"],"description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as...","inputPerM":5,"outputPerM":15,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":15,"cacheReadPerM":null}]},{"id":"openai/gpt-4o-2024-08-06","name":"GPT-4o (2024-08-06)","vendor":"openai","releasedAt":"2024-08-06","contextLength":128000,"maxOutput":16384,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools","web_search_options"],"description":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format. Read more [here](https://openai.com/index/introducing-structured-outputs-in-the-api/). GPT-4o (\"o\" for \"omni\") is...","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":1.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":1.25}]},{"id":"openai/gpt-4o-2024-11-20","name":"GPT-4o (2024-11-20)","vendor":"openai","releasedAt":"2024-11-20","contextLength":128000,"maxOutput":16384,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools","web_search_options"],"description":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability. It’s also better at working with uploaded...","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":1.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":1.25}]},{"id":"openai/gpt-4o-mini","name":"GPT-4o-mini","vendor":"openai","releasedAt":"2024-07-18","contextLength":128000,"maxOutput":16384,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools","web_search_options"],"description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.075,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.075}]},{"id":"openai/gpt-4o-mini-2024-07-18","name":"GPT-4o-mini (2024-07-18)","vendor":"openai","releasedAt":"2024-07-18","contextLength":128000,"maxOutput":16384,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools","web_search_options"],"description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.075,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.075}]},{"id":"openai/gpt-4-turbo","name":"GPT-4 Turbo","vendor":"openai","releasedAt":"2024-04-09","contextLength":128000,"maxOutput":4096,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling.\n\nTraining data: up to December 2023.","inputPerM":10,"outputPerM":30,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":10,"outputPerM":30,"cacheReadPerM":null}]},{"id":"openai/gpt-5","name":"GPT-5","vendor":"openai","releasedAt":"2025-08-07","contextLength":400000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy...","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125}]},{"id":"openai/gpt-5.1","name":"GPT-5.1","vendor":"openai","releasedAt":"2025-11-13","contextLength":400000,"maxOutput":128000,"input":["image","text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning...","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125}]},{"id":"openai/gpt-5.1-codex","name":"GPT-5.1-Codex","vendor":"openai","releasedAt":"2025-11-13","contextLength":400000,"maxOutput":128000,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks....","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.13,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.13}]},{"id":"openai/gpt-5.1-codex-max","name":"GPT-5.1-Codex-Max","vendor":"openai","releasedAt":"2025-12-04","contextLength":400000,"maxOutput":128000,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks. It is based on an updated version of the 5.1 reasoning stack and trained on agentic...","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":10,"cacheReadPerM":0.125}]},{"id":"openai/gpt-5.1-codex-mini","name":"GPT-5.1-Codex-Mini","vendor":"openai","releasedAt":"2025-11-13","contextLength":400000,"maxOutput":128000,"input":["image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":0.03}]},{"id":"openai/gpt-5.2","name":"GPT-5.2","vendor":"openai","releasedAt":"2025-12-10","contextLength":400000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1. It uses adaptive reasoning to allocate computation dynamically, responding quickly...","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175}]},{"id":"openai/gpt-5.2-chat","name":"GPT-5.2 Chat","vendor":"openai","releasedAt":"2025-12-10","contextLength":128000,"maxOutput":32000,"input":["file","image","text"],"output":["text"],"features":["structured_outputs","tools"],"description":"GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligence. It uses adaptive reasoning to selectively “think” on...","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175}]},{"id":"openai/gpt-5.2-codex","name":"GPT-5.2-Codex","vendor":"openai","releasedAt":"2026-01-14","contextLength":400000,"maxOutput":128000,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks....","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175}]},{"id":"openai/gpt-5.2-pro","name":"GPT-5.2 Pro","vendor":"openai","releasedAt":"2025-12-10","contextLength":400000,"maxOutput":128000,"input":["image","text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro. It is optimized for complex tasks that require step-by-step reasoning,...","inputPerM":21,"outputPerM":168,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":21,"outputPerM":168,"cacheReadPerM":null}]},{"id":"openai/gpt-5.3-codex","name":"GPT-5.3-Codex","vendor":"openai","releasedAt":"2026-02-24","contextLength":400000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2. It achieves state-of-the-art results...","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.75,"outputPerM":14,"cacheReadPerM":0.175}]},{"id":"openai/gpt-5.4","name":"GPT-5.4","vendor":"openai","releasedAt":"2026-03-05","contextLength":1050000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system. It features a 1M+ token context window (922K input, 128K output) with support for...","inputPerM":2.5,"outputPerM":15,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":15,"cacheReadPerM":0.25}]},{"id":"openai/gpt-5.4-image-2","name":"GPT-5.4 Image 2","vendor":"openai","releasedAt":"2026-04-21","contextLength":272000,"maxOutput":128000,"input":["image","text","file"],"output":["image","text"],"features":["reasoning","structured_outputs"],"description":"[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2. It enables rich multimodal workflows, allowing users to seamlessly move between reasoning, coding, and...","inputPerM":8,"outputPerM":15,"cacheReadPerM":2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":8,"outputPerM":15,"cacheReadPerM":2}]},{"id":"openai/gpt-5.4-mini","name":"GPT-5.4 Mini","vendor":"openai","releasedAt":"2026-03-17","contextLength":400000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads. It supports text and image inputs with strong performance across reasoning, coding,...","inputPerM":0.75,"outputPerM":4.5,"cacheReadPerM":0.075,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.75,"outputPerM":4.5,"cacheReadPerM":0.075}]},{"id":"openai/gpt-5.4-nano","name":"GPT-5.4 Nano","vendor":"openai","releasedAt":"2026-03-17","contextLength":400000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks. It supports text and image inputs and is designed for low-latency...","inputPerM":0.2,"outputPerM":1.25,"cacheReadPerM":0.02,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":1.25,"cacheReadPerM":0.02}]},{"id":"openai/gpt-5.4-pro","name":"GPT-5.4 Pro","vendor":"openai","releasedAt":"2026-03-05","contextLength":1050000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks. It features a 1M+ token context window (922K input, 128K...","inputPerM":30,"outputPerM":180,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":30,"outputPerM":180,"cacheReadPerM":null}]},{"id":"openai/gpt-5.5","name":"GPT-5.5","vendor":"openai","releasedAt":"2026-04-24","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5}]},{"id":"openai/gpt-5.5-pro","name":"GPT-5.5 Pro","vendor":"openai","releasedAt":"2026-04-24","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads. It features a 1M+ token context window (922K input, 128K output) with support for...","inputPerM":30,"outputPerM":180,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":30,"outputPerM":180,"cacheReadPerM":null}]},{"id":"openai/gpt-5.6-luna","name":"GPT-5.6 Luna","vendor":"openai","releasedAt":"2026-07-09","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for...","inputPerM":0.2,"outputPerM":1.2,"cacheReadPerM":0.02,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":1.2,"cacheReadPerM":0.02}]},{"id":"openai/gpt-5.6-luna-pro","name":"GPT-5.6 Luna Pro","vendor":"openai","releasedAt":"2026-07-09","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.6 Luna Pro is the same underlying model as [GPT-5.6 Luna](https://openrouter.ai/openai/gpt-5.6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":0.2,"outputPerM":1.2,"cacheReadPerM":0.02,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":1.2,"cacheReadPerM":0.02}]},{"id":"openai/gpt-5.6-sol","name":"GPT-5.6 Sol","vendor":"openai","releasedAt":"2026-07-09","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.4,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.4}]},{"id":"openai/gpt-5.6-sol-pro","name":"GPT-5.6 Sol Pro","vendor":"openai","releasedAt":"2026-07-09","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.6 Sol Pro is the same underlying model as [GPT-5.6 Sol](https://openrouter.ai/openai/gpt-5.6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.4,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":4,"outputPerM":20,"cacheReadPerM":0.4}]},{"id":"openai/gpt-5.6-terra","name":"GPT-5.6 Terra","vendor":"openai","releasedAt":"2026-07-09","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier. It is suited for everyday coding, reasoning, and agentic...","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2}]},{"id":"openai/gpt-5.6-terra-pro","name":"GPT-5.6 Terra Pro","vendor":"openai","releasedAt":"2026-07-09","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5.6 Terra Pro is the same underlying model as [GPT-5.6 Terra](https://openrouter.ai/openai/gpt-5.6-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":12,"cacheReadPerM":0.2}]},{"id":"openai/gpt-5-image","name":"GPT-5 Image","vendor":"openai","releasedAt":"2025-10-14","contextLength":400000,"maxOutput":128000,"input":["image","text","file"],"output":["image","text"],"features":["reasoning","structured_outputs"],"description":"[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities. It offers major improvements in reasoning, code quality, and user experience while incorporating GPT Image 1's superior instruction following,...","inputPerM":10,"outputPerM":10,"cacheReadPerM":1.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":10,"outputPerM":10,"cacheReadPerM":1.25}]},{"id":"openai/gpt-5-image-mini","name":"GPT-5 Image Mini","vendor":"openai","releasedAt":"2025-10-16","contextLength":400000,"maxOutput":128000,"input":["file","image","text"],"output":["image","text"],"features":["reasoning","structured_outputs"],"description":"GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation. This natively multimodal model features superior instruction following, text...","inputPerM":2.5,"outputPerM":2,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":2,"cacheReadPerM":0.25}]},{"id":"openai/gpt-5-mini","name":"GPT-5 Mini","vendor":"openai","releasedAt":"2025-08-07","contextLength":400000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost....","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":2,"cacheReadPerM":0.025}]},{"id":"openai/gpt-5-nano","name":"GPT-5 Nano","vendor":"openai","releasedAt":"2025-08-07","contextLength":400000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments. While limited in reasoning depth compared to its larger...","inputPerM":0.05,"outputPerM":0.4,"cacheReadPerM":0.005,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.4,"cacheReadPerM":0.005}]},{"id":"openai/gpt-5-pro","name":"GPT-5 Pro","vendor":"openai","releasedAt":"2025-10-06","contextLength":400000,"maxOutput":128000,"input":["image","text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and...","inputPerM":15,"outputPerM":120,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":15,"outputPerM":120,"cacheReadPerM":null}]},{"id":"openai/gpt-6.1-sol","name":"GPT-6.1 Sol","vendor":"openai","releasedAt":"2026-09-29","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1}]},{"id":"openai/gpt-6.1-sol-pro","name":"GPT-6.1 Sol Pro","vendor":"openai","releasedAt":"2026-09-29","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6.1 Sol Pro is the same underlying model as [GPT-6.1 Sol](https://openrouter.ai/openai/gpt-6.1-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.1}]},{"id":"openai/gpt-6.1-terra","name":"GPT-6.1 Terra","vendor":"openai","releasedAt":"2026-10-26","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots...","inputPerM":1,"outputPerM":5,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":5,"cacheReadPerM":0.05}]},{"id":"openai/gpt-6.1-terra-pro","name":"GPT-6.1 Terra Pro","vendor":"openai","releasedAt":"2026-10-26","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6.1 Terra Pro is the same underlying model as [GPT-6.1 Terra](https://openrouter.ai/openai/gpt-6.1-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":1,"outputPerM":5,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":5,"cacheReadPerM":0.05}]},{"id":"openai/gpt-6-astra","name":"GPT-6 Astra","vendor":"openai","releasedAt":"2026-09-04","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...","inputPerM":10,"outputPerM":50,"cacheReadPerM":1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":10,"outputPerM":50,"cacheReadPerM":1}]},{"id":"openai/gpt-6-astra-pro","name":"GPT-6 Astra Pro","vendor":"openai","releasedAt":"2026-09-04","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6 Astra Pro is the same underlying model as [GPT-6 Astra](https://openrouter.ai/openai/gpt-6-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":10,"outputPerM":50,"cacheReadPerM":1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":10,"outputPerM":50,"cacheReadPerM":1}]},{"id":"openai/gpt-6-luna","name":"GPT-6 Luna","vendor":"openai","releasedAt":"2026-09-22","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...","inputPerM":0.1,"outputPerM":0.5,"cacheReadPerM":0.01,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.5,"cacheReadPerM":0.01}]},{"id":"openai/gpt-6-luna-pro","name":"GPT-6 Luna Pro","vendor":"openai","releasedAt":"2026-09-22","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6 Luna Pro is the same underlying model as [GPT-6 Luna](https://openrouter.ai/openai/gpt-6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":0.1,"outputPerM":0.5,"cacheReadPerM":0.01,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.5,"cacheReadPerM":0.01}]},{"id":"openai/gpt-6-sol","name":"GPT-6 Sol","vendor":"openai","releasedAt":"2026-09-22","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2}]},{"id":"openai/gpt-6-sol-pro","name":"GPT-6 Sol Pro","vendor":"openai","releasedAt":"2026-09-22","contextLength":1050000,"maxOutput":128000,"input":["file","image","text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GPT-6 Sol Pro is the same underlying model as [GPT-6 Sol](https://openrouter.ai/openai/gpt-6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":10,"cacheReadPerM":0.2}]},{"id":"openai/gpt-audio","name":"GPT Audio","vendor":"openai","releasedAt":"2026-01-19","contextLength":128000,"maxOutput":16384,"input":["text","audio"],"output":["text","audio"],"features":["structured_outputs","tools"],"description":"The gpt-audio model is OpenAI's first generally available audio model. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Audio is priced...","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":10,"cacheReadPerM":null}]},{"id":"openai/gpt-audio-mini","name":"GPT Audio Mini","vendor":"openai","releasedAt":"2026-01-19","contextLength":128000,"maxOutput":16384,"input":["text","audio"],"output":["text","audio"],"features":["structured_outputs","tools"],"description":"A cost-efficient version of GPT Audio. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Input is priced at $0.60 per million...","inputPerM":0.6,"outputPerM":2.4,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":2.4,"cacheReadPerM":null}]},{"id":"openai/gpt-chat-latest","name":"GPT Chat Latest","vendor":"openai","releasedAt":"2026-05-05","contextLength":400000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["structured_outputs","tools"],"description":"GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT. As OpenAI rolls out new Instant model updates...","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5}]},{"id":"openai/gpt-oss-120b","name":"gpt-oss-120b","vendor":"openai","releasedAt":"2025-08-05","contextLength":131072,"maxOutput":65536,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized...","inputPerM":0.014,"outputPerM":0.051,"cacheReadPerM":0.012,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-01","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.075},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.022,"outputPerM":0.082,"cacheReadPerM":null},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.017,"outputPerM":0.061,"cacheReadPerM":null},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.016,"outputPerM":0.059,"cacheReadPerM":null},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.029,"outputPerM":0.108,"cacheReadPerM":0.026},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.029,"outputPerM":0.107,"cacheReadPerM":0.026},{"from":"2026-11-02","to":"2026-11-03","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.075},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.022,"outputPerM":0.081,"cacheReadPerM":0.02},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.028,"outputPerM":0.103,"cacheReadPerM":0.025},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.026,"outputPerM":0.096,"cacheReadPerM":0.023},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.024,"outputPerM":0.089,"cacheReadPerM":0.022},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.011,"outputPerM":0.042,"cacheReadPerM":0.01},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.019,"outputPerM":0.069,"cacheReadPerM":0.017},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.075},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.029,"outputPerM":0.106,"cacheReadPerM":0.026},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.029,"outputPerM":0.109,"cacheReadPerM":0.026},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.014,"outputPerM":0.051,"cacheReadPerM":0.012}]},{"id":"openai/gpt-oss-20b","name":"gpt-oss-20b","vendor":"openai","releasedAt":"2025-08-05","contextLength":131072,"maxOutput":117964,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...","inputPerM":0.03,"outputPerM":0.13,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-01","inputPerM":0.018,"outputPerM":0.09,"cacheReadPerM":0.009},{"from":"2026-11-01","to":"2026-11-03","inputPerM":0.03,"outputPerM":0.13,"cacheReadPerM":0.03},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.018,"outputPerM":0.09,"cacheReadPerM":0.009},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.03,"outputPerM":0.13,"cacheReadPerM":0.03},{"from":"2026-11-04","to":"2026-11-05","inputPerM":0.018,"outputPerM":0.09,"cacheReadPerM":0.009},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.03,"outputPerM":0.13,"cacheReadPerM":0.03}]},{"id":"openai/gpt-oss-safeguard-20b","name":"gpt-oss-safeguard-20b","vendor":"openai","releasedAt":"2025-10-29","contextLength":131072,"maxOutput":65536,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0375,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.075,"outputPerM":0.3,"cacheReadPerM":0.0375}]},{"id":"openai/o1","name":"o1","vendor":"openai","releasedAt":"2024-12-17","contextLength":200000,"maxOutput":100000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding. The o1 model series is trained with large-scale reinforcement learning to reason...","inputPerM":15,"outputPerM":60,"cacheReadPerM":7.5,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":15,"outputPerM":60,"cacheReadPerM":7.5}]},{"id":"openai/o1-pro","name":"o1-pro","vendor":"openai","releasedAt":"2025-03-19","contextLength":200000,"maxOutput":100000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs"],"description":"The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o1-pro model uses more compute to think harder and provide...","inputPerM":150,"outputPerM":600,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":150,"outputPerM":600,"cacheReadPerM":null}]},{"id":"openai/o3","name":"o3","vendor":"openai","releasedAt":"2025-04-16","contextLength":200000,"maxOutput":100000,"input":["image","text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following....","inputPerM":2,"outputPerM":8,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":8,"cacheReadPerM":0.5}]},{"id":"openai/o3-mini","name":"o3 Mini","vendor":"openai","releasedAt":"2025-01-31","contextLength":200000,"maxOutput":100000,"input":["text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding. This model supports the `reasoning_effort` parameter, which can be set to...","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.55,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.55}]},{"id":"openai/o3-mini-high","name":"o3 Mini High","vendor":"openai","releasedAt":"2025-02-12","contextLength":200000,"maxOutput":100000,"input":["text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high. o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and...","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.55,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.55}]},{"id":"openai/o3-pro","name":"o3 Pro","vendor":"openai","releasedAt":"2025-06-10","contextLength":200000,"maxOutput":100000,"input":["text","file","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently...","inputPerM":20,"outputPerM":80,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":20,"outputPerM":80,"cacheReadPerM":null}]},{"id":"openai/o4-mini","name":"o4 Mini","vendor":"openai","releasedAt":"2025-04-16","contextLength":200000,"maxOutput":100000,"input":["image","text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning...","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.275,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.275}]},{"id":"openai/o4-mini-high","name":"o4 Mini High","vendor":"openai","releasedAt":"2025-04-16","contextLength":200000,"maxOutput":100000,"input":["image","text","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high. OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining...","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.275,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":1.1,"outputPerM":4.4,"cacheReadPerM":0.275}]},{"id":"particle-ai/particle-1","name":"Particle-1","vendor":"particle-ai","releasedAt":"2026-10-28","contextLength":1048576,"maxOutput":131072,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as terminal workflows, codebase changes, and data...","inputPerM":0.05,"outputPerM":0.5,"cacheReadPerM":0.003,"firstSeen":"2026-10-29","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-29","to":"2026-10-30","inputPerM":0.25,"outputPerM":1,"cacheReadPerM":0.03},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.25,"outputPerM":1,"cacheReadPerM":0.003},{"from":"2026-10-30","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.5,"cacheReadPerM":0.003}]},{"id":"perceptron/perceptron-mk1","name":"Perceptron Mk1","vendor":"perceptron","releasedAt":"2026-05-12","contextLength":32768,"maxOutput":8192,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs"],"description":"Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces detailed visual understanding...","inputPerM":0.15,"outputPerM":1.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":1.5,"cacheReadPerM":null}]},{"id":"perceptron/perceptron-mk1.5","name":"Perceptron Mk1.5","vendor":"perceptron","releasedAt":"2026-09-25","contextLength":36864,"maxOutput":8192,"input":["text","image","video","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Perceptron Mk1.5 is Perceptron's embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers with text plus optional structured annotations: points, boxes, polygons, tracks,...","inputPerM":0.15,"outputPerM":1.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":1.5,"cacheReadPerM":null}]},{"id":"perplexity/sonar","name":"Sonar","vendor":"perplexity","releasedAt":"2025-01-27","contextLength":127072,"maxOutput":114364,"input":["text","image"],"output":["text"],"features":["web_search_options"],"description":"Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources. It is designed for companies seeking to integrate lightweight question-and-answer features...","inputPerM":1,"outputPerM":1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":1,"cacheReadPerM":null}]},{"id":"perplexity/sonar-deep-research","name":"Sonar Deep Research","vendor":"perplexity","releasedAt":"2025-03-07","contextLength":128000,"maxOutput":115200,"input":["text"],"output":["text"],"features":["reasoning","web_search_options"],"description":"Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics. It autonomously searches, reads, and evaluates sources, refining its approach as it gathers...","inputPerM":2,"outputPerM":8,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":8,"cacheReadPerM":null}]},{"id":"perplexity/sonar-pro","name":"Sonar Pro","vendor":"perplexity","releasedAt":"2025-03-07","contextLength":200000,"maxOutput":8000,"input":["text","image"],"output":["text"],"features":["web_search_options"],"description":"Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) For enterprises seeking more advanced capabilities, the Sonar Pro API can handle in-depth, multi-step queries with added extensibility, like...","inputPerM":3,"outputPerM":15,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":15,"cacheReadPerM":null}]},{"id":"perplexity/sonar-pro-search","name":"Sonar Pro Search","vendor":"perplexity","releasedAt":"2025-10-30","contextLength":200000,"maxOutput":8000,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","web_search_options"],"description":"Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system. It is designed for deeper reasoning and analysis. Pricing is based...","inputPerM":3,"outputPerM":15,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":3,"outputPerM":15,"cacheReadPerM":null}]},{"id":"perplexity/sonar-reasoning-pro","name":"Sonar Reasoning Pro","vendor":"perplexity","releasedAt":"2025-03-07","contextLength":128000,"maxOutput":115200,"input":["text","image"],"output":["text"],"features":["reasoning","web_search_options"],"description":"Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) Sonar Reasoning Pro is a premier reasoning model powered by DeepSeek R1 with Chain of Thought (CoT). Designed for...","inputPerM":2,"outputPerM":8,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":8,"cacheReadPerM":null}]},{"id":"poolside/laguna-s-2.1","name":"Laguna S 2.1","vendor":"poolside","releasedAt":"2026-07-21","contextLength":1048576,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Laguna S 2.1 is the latest coding agent model from [Poolside](<https://poolside.ai/>). Laguna S 2.1 is a 118B total parameter model with 8B active parameters, scoring 70.2% on Terminal-Bench 2.1 and...","inputPerM":0.09,"outputPerM":0.18,"cacheReadPerM":0.009,"firstSeen":"2026-10-28","lastSeen":"2026-10-31","gone":"2026-10-31","history":[{"from":"2026-10-28","to":"2026-10-31","inputPerM":0.09,"outputPerM":0.18,"cacheReadPerM":0.009}]},{"id":"poolside/laguna-xs-2.1","name":"Laguna XS 2.1","vendor":"poolside","releasedAt":"2026-07-02","contextLength":262144,"maxOutput":32768,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Laguna XS 2.1 is the latest coding agent model in the 33B-A3B category from [Poolside](https://poolside.ai/) and a step forward from their Laguna XS.2 model (released in April 2026). It combines...","inputPerM":0.06,"outputPerM":0.12,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-10-31","gone":"2026-10-31","history":[{"from":"2026-10-28","to":"2026-10-31","inputPerM":0.06,"outputPerM":0.12,"cacheReadPerM":0.03}]},{"id":"prism-ml/ternary-bonsai-2-27b","name":"Ternary Bonsai 2 27B","vendor":"prism-ml","releasedAt":"2026-09-18","contextLength":262144,"maxOutput":32768,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window. Ternary compression shrinks...","inputPerM":0.075,"outputPerM":0.5,"cacheReadPerM":0.0375,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.075,"outputPerM":0.5,"cacheReadPerM":0.0375}]},{"id":"prism-ml/ternary-bonsai-2-9b","name":"Ternary Bonsai 2 9B","vendor":"prism-ml","releasedAt":"2026-10-16","contextLength":262144,"maxOutput":65536,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Ternary Bonsai 2 9B is PrismML’s compact, open-weight language model for conversational AI, reasoning, and tool use. Built on Qwen3.5, it uses ternary weights packed into 2 bits for efficient...","inputPerM":0.08,"outputPerM":0.13,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.08,"outputPerM":0.13,"cacheReadPerM":0.04}]},{"id":"qwen/qwen-2.5-72b-instruct","name":"Qwen2.5 72B Instruct","vendor":"qwen","releasedAt":"2024-09-19","contextLength":32768,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...","inputPerM":0.36,"outputPerM":0.4,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.36,"outputPerM":0.4,"cacheReadPerM":null}]},{"id":"qwen/qwen-2.5-7b-instruct","name":"Qwen2.5 7B Instruct","vendor":"qwen","releasedAt":"2024-10-16","contextLength":32768,"maxOutput":29491,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen2.5 7B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":null}]},{"id":"qwen/qwen-2.5-coder-32b-instruct","name":"Qwen2.5 Coder 32B Instruct","vendor":"qwen","releasedAt":"2024-11-11","contextLength":32768,"maxOutput":29491,"input":["text"],"output":["text"],"features":[],"description":"Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). Qwen2.5-Coder brings the following improvements upon CodeQwen1.5: - Significantly improvements in **code generation**, **code reasoning**...","inputPerM":0.66,"outputPerM":1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.66,"outputPerM":1,"cacheReadPerM":null}]},{"id":"qwen/qwen2.5-vl-72b-instruct","name":"Qwen2.5 VL 72B Instruct","vendor":"qwen","releasedAt":"2025-02-01","contextLength":128000,"maxOutput":115200,"input":["text","image"],"output":["text"],"features":["structured_outputs"],"description":"Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.","inputPerM":0.8,"outputPerM":1,"cacheReadPerM":0.4,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.8,"outputPerM":1,"cacheReadPerM":0.4}]},{"id":"qwen/qwen3-14b","name":"Qwen3 14B","vendor":"qwen","releasedAt":"2025-04-28","contextLength":40960,"maxOutput":16384,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It supports seamless switching between a \"thinking\" mode for...","inputPerM":0.12,"outputPerM":0.24,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.12,"outputPerM":0.24,"cacheReadPerM":null}]},{"id":"qwen/qwen3-235b-a22b-2507","name":"Qwen3 235B A22B Instruct 2507","vendor":"qwen","releasedAt":"2025-07-21","contextLength":262144,"maxOutput":117964,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following,...","inputPerM":0.14,"outputPerM":0.8,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-03","inputPerM":0.09,"outputPerM":0.55,"cacheReadPerM":null},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.14,"outputPerM":0.8,"cacheReadPerM":0.05},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.09,"outputPerM":0.55,"cacheReadPerM":null},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.14,"outputPerM":0.8,"cacheReadPerM":0.05},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.09,"outputPerM":0.55,"cacheReadPerM":null},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.14,"outputPerM":0.8,"cacheReadPerM":0.05},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.09,"outputPerM":0.55,"cacheReadPerM":null},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.14,"outputPerM":0.8,"cacheReadPerM":0.05}]},{"id":"qwen/qwen3-235b-a22b-thinking-2507","name":"Qwen3 235B A22B Thinking 2507","vendor":"qwen","releasedAt":"2025-07-25","contextLength":128000,"maxOutput":16384,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...","inputPerM":0.45,"outputPerM":3.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.45,"outputPerM":3.5,"cacheReadPerM":null}]},{"id":"qwen/qwen3-30b-a3b","name":"Qwen3 30B A3B","vendor":"qwen","releasedAt":"2025-04-28","contextLength":40960,"maxOutput":16384,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks. Its unique...","inputPerM":0.12,"outputPerM":0.5,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.12,"outputPerM":0.5,"cacheReadPerM":null}]},{"id":"qwen/qwen3-30b-a3b-instruct-2507","name":"Qwen3 30B A3B Instruct 2507","vendor":"qwen","releasedAt":"2025-07-29","contextLength":262144,"maxOutput":235929,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and...","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":0.1}]},{"id":"qwen/qwen3-32b","name":"Qwen3 32B","vendor":"qwen","releasedAt":"2025-04-28","contextLength":131072,"maxOutput":16384,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a \"thinking\" mode for...","inputPerM":0.08,"outputPerM":0.28,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.08,"outputPerM":0.28,"cacheReadPerM":null}]},{"id":"qwen/qwen3.5-122b-a10b","name":"Qwen3.5-122B-A10B","vendor":"qwen","releasedAt":"2026-02-25","contextLength":262144,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of...","inputPerM":0.26,"outputPerM":2.08,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.26,"outputPerM":2.08,"cacheReadPerM":null}]},{"id":"qwen/qwen3.5-27b","name":"Qwen3.5-27B","vendor":"qwen","releasedAt":"2026-02-25","contextLength":262144,"maxOutput":81920,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance. Its overall capabilities are comparable to those of...","inputPerM":0.26,"outputPerM":2.6,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.26,"outputPerM":2.6,"cacheReadPerM":null}]},{"id":"qwen/qwen3.5-35b-a3b","name":"Qwen3.5-35B-A3B","vendor":"qwen","releasedAt":"2026-02-25","contextLength":262144,"maxOutput":81920,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall...","inputPerM":0.14,"outputPerM":1,"cacheReadPerM":0.05,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.14,"outputPerM":1,"cacheReadPerM":0.05}]},{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen3.5 397B A17B","vendor":"qwen","releasedAt":"2026-02-16","contextLength":262144,"maxOutput":235929,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers...","inputPerM":0.5,"outputPerM":3.6,"cacheReadPerM":0.3,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.5,"outputPerM":3.6,"cacheReadPerM":0.3}]},{"id":"qwen/qwen3.5-9b","name":"Qwen3.5-9B","vendor":"qwen","releasedAt":"2026-03-10","contextLength":262144,"maxOutput":235929,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture. It uses a unified vision-language design...","inputPerM":0.17,"outputPerM":0.25,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.17,"outputPerM":0.25,"cacheReadPerM":null}]},{"id":"qwen/qwen3.5-flash-02-23","name":"Qwen3.5-Flash","vendor":"qwen","releasedAt":"2026-02-25","contextLength":1000000,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. Compared to the...","inputPerM":0.065,"outputPerM":0.26,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.065,"outputPerM":0.26,"cacheReadPerM":null}]},{"id":"qwen/qwen3.5-plus-02-15","name":"Qwen3.5 Plus 2026-02-15","vendor":"qwen","releasedAt":"2026-02-16","contextLength":1000000,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixture-of-experts models, achieving higher inference efficiency. In a variety of...","inputPerM":0.26,"outputPerM":1.56,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.26,"outputPerM":1.56,"cacheReadPerM":null}]},{"id":"qwen/qwen3.5-plus-20260420","name":"Qwen3.5 Plus 2026-04-20","vendor":"qwen","releasedAt":"2026-04-27","contextLength":1000000,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba. It accepts text, image, and video input and produces text output, with a 1M token context window. This...","inputPerM":0.3,"outputPerM":1.8,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":1.8,"cacheReadPerM":null}]},{"id":"qwen/qwen3.6-27b","name":"Qwen3.6 27B","vendor":"qwen","releasedAt":"2026-04-27","contextLength":262144,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026. It features hybrid multimodal capabilities — accepting text, image, and video inputs...","inputPerM":0.45,"outputPerM":2.7,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-29","inputPerM":0.45,"outputPerM":2.7,"cacheReadPerM":null},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.32,"outputPerM":3.2,"cacheReadPerM":null},{"from":"2026-10-31","to":"2026-11-06","inputPerM":0.45,"outputPerM":2.7,"cacheReadPerM":null}]},{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen3.6 35B A3B","vendor":"qwen","releasedAt":"2026-04-27","contextLength":262144,"maxOutput":235929,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated...","inputPerM":0.1,"outputPerM":0.9,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.9,"cacheReadPerM":0.025}]},{"id":"qwen/qwen3.6-flash","name":"Qwen3.6 Flash","vendor":"qwen","releasedAt":"2026-04-27","contextLength":1000000,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series. It supports text, image, and video input with a 1M token context window. Tiered pricing kicks in...","inputPerM":0.1875,"outputPerM":1.125,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1875,"outputPerM":1.125,"cacheReadPerM":null}]},{"id":"qwen/qwen3.6-plus","name":"Qwen3.6 Plus","vendor":"qwen","releasedAt":"2026-04-02","contextLength":1000000,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference. Compared to the 3.5 series, it delivers...","inputPerM":0.325,"outputPerM":1.95,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.325,"outputPerM":1.95,"cacheReadPerM":null}]},{"id":"qwen/qwen3.7-flash","name":"Qwen3.7 Flash","vendor":"qwen","releasedAt":"2026-07-27","contextLength":1000000,"maxOutput":65536,"input":["text","image","video"],"output":["text"],"features":["reasoning","tools"],"description":"Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer interaction, with strengths in object recognition, spatial understanding, and real-world...","inputPerM":0.03,"outputPerM":0.13,"cacheReadPerM":0.006,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.03,"outputPerM":0.13,"cacheReadPerM":0.006}]},{"id":"qwen/qwen3.7-max","name":"Qwen3.7 Max","vendor":"qwen","releasedAt":"2026-05-21","contextLength":1000000,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks,...","inputPerM":1.475,"outputPerM":4.425,"cacheReadPerM":0.295,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.475,"outputPerM":4.425,"cacheReadPerM":0.295}]},{"id":"qwen/qwen3.7-plus","name":"Qwen3.7 Plus","vendor":"qwen","releasedAt":"2026-06-03","contextLength":1000000,"maxOutput":131072,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its...","inputPerM":0.32,"outputPerM":1.28,"cacheReadPerM":0.064,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.32,"outputPerM":1.28,"cacheReadPerM":0.064}]},{"id":"qwen/qwen3.8-2.4t-a95b","name":"Qwen3.8 2.4T A95B","vendor":"qwen","releasedAt":"2026-08-12","contextLength":1048576,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is...","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.25}]},{"id":"qwen/qwen3.8-27b","name":"Qwen3.8 27B","vendor":"qwen","releasedAt":"2026-08-14","contextLength":1024000,"maxOutput":235929,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...","inputPerM":0.15,"outputPerM":1.875,"cacheReadPerM":0.0375,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-30","inputPerM":0.039,"outputPerM":1.25,"cacheReadPerM":0.0039},{"from":"2026-10-30","to":"2026-10-31","inputPerM":0.15,"outputPerM":1.875,"cacheReadPerM":0.0375},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.04,"outputPerM":2.3,"cacheReadPerM":0.02},{"from":"2026-11-01","to":"2026-11-06","inputPerM":0.15,"outputPerM":1.875,"cacheReadPerM":0.0375}]},{"id":"qwen/qwen3.8-flash","name":"Qwen3.8 Flash","vendor":"qwen","releasedAt":"2026-08-26","contextLength":1000000,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.8 Flash is a multimodal reasoning model from Alibaba. It is suited for coding assistance, agentic workflows, visual understanding, document and codebase analysis, desktop interaction, chart analysis, and long-video analysis.","inputPerM":0.15,"outputPerM":0.47,"cacheReadPerM":0.016,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.47,"cacheReadPerM":0.016}]},{"id":"qwen/qwen3.8-max-0902","name":"Qwen3.8 Max (0902)","vendor":"qwen","releasedAt":"2026-09-03","contextLength":1000000,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text,...","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.25}]},{"id":"qwen/qwen3.8-max-prime","name":"Qwen3.8 Max Prime","vendor":"qwen","releasedAt":"2026-09-23","contextLength":1000000,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.8 Max Prime is a higher-throughput variant of Qwen3.8 Max from Alibaba's Qwen team, served as a separate SKU at a higher price point. It accepts text, image, and video...","inputPerM":4,"outputPerM":12,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":4,"outputPerM":12,"cacheReadPerM":0.5}]},{"id":"qwen/qwen3.8-omni-flash","name":"Qwen3.8 Omni Flash","vendor":"qwen","releasedAt":"2026-09-21","contextLength":1000000,"maxOutput":131072,"input":["text","image","audio","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3.8 Omni Flash is an omni-modal reasoning model from Alibaba, the first Qwen model built around agentic capabilities with native audio-video understanding. It is suited for audio-video analysis and summarization,...","inputPerM":0.15,"outputPerM":0.47,"cacheReadPerM":0.016,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.47,"cacheReadPerM":0.016}]},{"id":"qwen/qwen3-coder","name":"Qwen3 Coder 480B A35B","vendor":"qwen","releasedAt":"2025-07-23","contextLength":262144,"maxOutput":65536,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over...","inputPerM":0.3,"outputPerM":1,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":1,"cacheReadPerM":0.1}]},{"id":"qwen/qwen3-coder-30b-a3b-instruct","name":"Qwen3 Coder 30B A3B Instruct","vendor":"qwen","releasedAt":"2025-07-31","contextLength":262144,"maxOutput":235929,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-01","inputPerM":0.07,"outputPerM":0.28,"cacheReadPerM":null},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":null},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.07,"outputPerM":0.28,"cacheReadPerM":null},{"from":"2026-11-03","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":null}]},{"id":"qwen/qwen3-coder-flash","name":"Qwen3 Coder Flash","vendor":"qwen","releasedAt":"2025-09-17","contextLength":1000000,"maxOutput":65536,"input":["text"],"output":["text"],"features":["tools"],"description":"Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus. It is a powerful coding agent model specializing in autonomous programming via tool calling...","inputPerM":0.195,"outputPerM":0.975,"cacheReadPerM":0.039,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.195,"outputPerM":0.975,"cacheReadPerM":0.039}]},{"id":"qwen/qwen3-coder-next","name":"Qwen3 Coder Next","vendor":"qwen","releasedAt":"2026-02-04","contextLength":262144,"maxOutput":235929,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per...","inputPerM":0.12,"outputPerM":0.8,"cacheReadPerM":0.07,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.12,"outputPerM":0.8,"cacheReadPerM":0.07}]},{"id":"qwen/qwen3-next-80b-a3b-instruct","name":"Qwen3 Next 80B A3B Instruct","vendor":"qwen","releasedAt":"2025-09-11","contextLength":262144,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks across reasoning, code generation, knowledge QA, and multilingual...","inputPerM":0.09,"outputPerM":1.1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-02","inputPerM":0.09,"outputPerM":1.1,"cacheReadPerM":null},{"from":"2026-11-03","to":"2026-11-05","inputPerM":0.1,"outputPerM":1.1,"cacheReadPerM":0.07},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.09,"outputPerM":1.1,"cacheReadPerM":null}]},{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"Qwen3 Next 80B A3B Thinking","vendor":"qwen","releasedAt":"2025-09-11","contextLength":262144,"maxOutput":235929,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default. It’s designed for hard multi-step problems; math proofs, code synthesis/debugging, logic, and agentic...","inputPerM":0.15,"outputPerM":1.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-10-28","gone":"2026-10-28","history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.15,"outputPerM":1.2,"cacheReadPerM":null}]},{"id":"qwen/qwen3-vl-235b-a22b-instruct","name":"Qwen3 VL 235B A22B Instruct","vendor":"qwen","releasedAt":"2025-09-23","contextLength":262144,"maxOutput":32768,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video. The Instruct model targets general vision-language use (VQA, document parsing, chart/table...","inputPerM":0.21,"outputPerM":1.9,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.21,"outputPerM":1.9,"cacheReadPerM":0.1}]},{"id":"qwen/qwen3-vl-30b-a3b-instruct","name":"Qwen3 VL 30B A3B Instruct","vendor":"qwen","releasedAt":"2025-10-06","contextLength":262144,"maxOutput":16384,"input":["text","image"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Instruct variant optimizes instruction-following for general multimodal tasks. It excels in perception...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":null}]},{"id":"qwen/qwen3-vl-30b-a3b-thinking","name":"Qwen3 VL 30B A3B Thinking","vendor":"qwen","releasedAt":"2025-10-06","contextLength":262144,"maxOutput":235929,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Thinking variant enhances reasoning in STEM, math, and complex tasks. It excels...","inputPerM":0.29,"outputPerM":1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.29,"outputPerM":1,"cacheReadPerM":null}]},{"id":"qwen/qwen3-vl-8b-instruct","name":"Qwen3 VL 8B Instruct","vendor":"qwen","releasedAt":"2025-10-14","contextLength":262144,"maxOutput":235929,"input":["image","text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video. It features improved multimodal fusion with Interleaved-MRoPE for long-horizon...","inputPerM":0.25,"outputPerM":0.75,"cacheReadPerM":0.12,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.25,"outputPerM":0.75,"cacheReadPerM":0.12}]},{"id":"qwen/qwen-plus","name":"Qwen-Plus","vendor":"qwen","releasedAt":"2025-02-01","contextLength":1000000,"maxOutput":32768,"input":["text"],"output":["text"],"features":["structured_outputs","tools"],"description":"Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination.","inputPerM":0.26,"outputPerM":0.78,"cacheReadPerM":0.052,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.26,"outputPerM":0.78,"cacheReadPerM":0.052}]},{"id":"reflection/beam-1.0-501b-a23b","name":"Beam 1.0 501B A23B","vendor":"reflection","releasedAt":"2026-10-16","contextLength":1048576,"maxOutput":943718,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per token, and a...","inputPerM":0.3,"outputPerM":1,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-05","inputPerM":0.25,"outputPerM":1,"cacheReadPerM":0.05},{"from":"2026-11-05","to":"2026-11-06","inputPerM":0.3,"outputPerM":1,"cacheReadPerM":0.04}]},{"id":"rekaai/reka-edge","name":"Reka Edge","vendor":"rekaai","releasedAt":"2026-03-20","contextLength":16384,"maxOutput":14745,"input":["image","text","video"],"output":["text"],"features":["structured_outputs","tools"],"description":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs. This model is optimized specifically to deliver industry-leading performance in image understanding,...","inputPerM":0.1,"outputPerM":0.1,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.1,"cacheReadPerM":null}]},{"id":"rekaai/reka-flash-3","name":"Reka Flash 3","vendor":"rekaai","releasedAt":"2025-03-12","contextLength":65536,"maxOutput":58982,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs"],"description":"Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka. It excels at general chat, coding tasks, instruction-following, and function calling. Featuring a...","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.2,"cacheReadPerM":null}]},{"id":"relace/relace-apply-3","name":"Relace Apply 3","vendor":"relace","releasedAt":"2025-09-26","contextLength":256000,"maxOutput":128000,"input":["text"],"output":["text"],"features":[],"description":"Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files. It can apply updates from GPT-4o, Claude, and others into your files at...","inputPerM":0.85,"outputPerM":1.25,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.85,"outputPerM":1.25,"cacheReadPerM":null}]},{"id":"relace/relace-search","name":"Relace Search","vendor":"relace","releasedAt":"2025-12-08","contextLength":256000,"maxOutput":128000,"input":["text"],"output":["text"],"features":["tools"],"description":"The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request. In contrast to RAG, relace-search performs agentic...","inputPerM":1,"outputPerM":3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":3,"cacheReadPerM":null}]},{"id":"ricursiveai/ripple-65b-preview","name":"ricursiveAI/Ripple-65B-Preview","vendor":"ricursiveai","releasedAt":"2026-11-04","contextLength":1048576,"maxOutput":131072,"input":["text"],"output":["text"],"features":["tools"],"description":"Ripple-65B-Preview by ricursiveAI, a 65B model with native tool calling, served in BF16.","inputPerM":0.2,"outputPerM":0.6,"cacheReadPerM":0.05,"firstSeen":"2026-11-04","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-11-04","to":"2026-11-06","inputPerM":0.2,"outputPerM":0.6,"cacheReadPerM":0.05}]},{"id":"sakana/fugu-max","name":"Fugu Max","vendor":"sakana","releasedAt":"2026-09-11","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools","web_search_options"],"description":"Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.25}]},{"id":"sakana/fugu-ultra","name":"Fugu Ultra","vendor":"sakana","releasedAt":"2026-06-24","contextLength":1000000,"maxOutput":128000,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools","web_search_options"],"description":"Fugu Ultra is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5}]},{"id":"sakana/fugu-ultra-v2","name":"Fugu Ultra v2","vendor":"sakana","releasedAt":"2026-09-11","contextLength":1000000,"maxOutput":128000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools","web_search_options"],"description":"Fugu Ultra v2 is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to...","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":5,"outputPerM":30,"cacheReadPerM":0.5}]},{"id":"sakana/sakana-namazu","name":"Sakana Namazu","vendor":"sakana","releasedAt":"2026-08-11","contextLength":262144,"maxOutput":65536,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools","web_search_options"],"description":"Sakana Namazu is a Japanese-specialized reasoning model from Sakana AI, based on Kimi K2.6 with additional training for Japanese language and business contexts. It is suited for Japanese instruction following,...","inputPerM":0.95,"outputPerM":4,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.95,"outputPerM":4,"cacheReadPerM":0.15}]},{"id":"sao10k/l3.1-euryale-70b","name":"Llama 3.1 Euryale 70B v2.2","vendor":"sao10k","releasedAt":"2024-08-28","contextLength":131072,"maxOutput":16384,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k). It is the successor of [Euryale L3 70B v2.1](/models/sao10k/l3-euryale-70b).","inputPerM":0.85,"outputPerM":0.85,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.85,"outputPerM":0.85,"cacheReadPerM":null}]},{"id":"sao10k/l3-lunaris-8b","name":"Llama 3 8B Lunaris","vendor":"sao10k","releasedAt":"2024-08-13","contextLength":8192,"maxOutput":7372,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3. It's a strategic merge of multiple models, designed to balance creativity with improved logic and general knowledge....","inputPerM":0.04,"outputPerM":0.05,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.04,"outputPerM":0.05,"cacheReadPerM":null}]},{"id":"scaledown/extract","name":"Extract","vendor":"scaledown","releasedAt":"2026-10-13","contextLength":1000000,"maxOutput":900000,"input":["text","image","file"],"output":["text"],"features":[],"description":"Extract is a description-driven entity extraction model from ScaleDown. Instead of a fixed set of entity types, callers describe the fields they want in plain English through `provider.options.scaledown.entities`, and the...","inputPerM":0.04,"outputPerM":0,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.04,"outputPerM":0,"cacheReadPerM":null}]},{"id":"scaledown/summarize","name":"Summarize","vendor":"scaledown","releasedAt":"2026-10-13","contextLength":1000000,"maxOutput":4000,"input":["text","image","file"],"output":["text"],"features":[],"description":"Summarize is an abstractive summarization model from ScaleDown. It condenses long text and documents into a fluent, faithful rewrite rather than lifting sentences from the source, and is suited for...","inputPerM":0.04,"outputPerM":0,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.04,"outputPerM":0,"cacheReadPerM":null}]},{"id":"stealth/crouton-alpha","name":"Crouton Alpha","vendor":"stealth","releasedAt":"2026-11-03","contextLength":1048576,"maxOutput":943718,"input":["text","image","video","file","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Crouton Alpha is a lightweight model built for scalable coding, agentic, and multimodal workflows. It's ideal for executing agentic sub-tasks, tool use, and document analysis in high-volume, repeated work. Developers...","inputPerM":0,"outputPerM":0,"cacheReadPerM":null,"firstSeen":"2026-11-04","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-11-04","to":"2026-11-06","inputPerM":0,"outputPerM":0,"cacheReadPerM":null}]},{"id":"stepfun/step-3.5-flash","name":"Step 3.5 Flash","vendor":"stepfun","releasedAt":"2026-01-29","contextLength":262144,"maxOutput":65536,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.1,"outputPerM":0.3,"cacheReadPerM":null}]},{"id":"stepfun/step-3.7-flash","name":"Step 3.7 Flash","vendor":"stepfun","releasedAt":"2026-05-28","contextLength":262144,"maxOutput":230400,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters...","inputPerM":0.2,"outputPerM":1.15,"cacheReadPerM":0.04,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.2,"outputPerM":1.15,"cacheReadPerM":0.04}]},{"id":"stepfun/step-5-preview","name":"Step 5 Preview","vendor":"stepfun","releasedAt":"2026-10-08","contextLength":1048576,"maxOutput":64000,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional...","inputPerM":1,"outputPerM":2.7,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":2.7,"cacheReadPerM":0.1}]},{"id":"tencent/hunyuan-a13b-instruct","name":"Hunyuan A13B Instruct","vendor":"tencent","releasedAt":"2025-07-08","contextLength":131072,"maxOutput":117964,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs"],"description":"Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought. It offers competitive benchmark...","inputPerM":0.14,"outputPerM":0.57,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.14,"outputPerM":0.57,"cacheReadPerM":null}]},{"id":"tencent/hy3","name":"Hy3","vendor":"tencent","releasedAt":"2026-07-06","contextLength":262144,"maxOutput":128000,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.0825,"outputPerM":0.33,"cacheReadPerM":0.020625},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.132,"outputPerM":0.528,"cacheReadPerM":0.033}]},{"id":"tencent/hy3-preview","name":"Hy3 preview","vendor":"tencent","releasedAt":"2026-04-22","contextLength":262144,"maxOutput":235929,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to...","inputPerM":0.18,"outputPerM":0.6,"cacheReadPerM":0.06,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.18,"outputPerM":0.6,"cacheReadPerM":0.06}]},{"id":"tencent/hy4","name":"Hy4","vendor":"tencent","releasedAt":"2026-11-04","contextLength":1048576,"maxOutput":65536,"input":["text","image"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity...","inputPerM":0.417,"outputPerM":1.2505,"cacheReadPerM":0.021,"firstSeen":"2026-11-04","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-11-04","to":"2026-11-06","inputPerM":0.417,"outputPerM":1.2505,"cacheReadPerM":0.021}]},{"id":"tencent/hy4-preview","name":"Hy4 preview","vendor":"tencent","releasedAt":"2026-08-28","contextLength":1048576,"maxOutput":64000,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks that...","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042},{"from":"2026-11-05","to":"2026-11-05","inputPerM":0.7506,"outputPerM":2.2509,"cacheReadPerM":0.0378},{"from":"2026-11-06","to":"2026-11-06","inputPerM":0.834,"outputPerM":2.501,"cacheReadPerM":0.042}]},{"id":"tencent/hy-mt2-1.8b","name":"Hy-MT2-1.8B","vendor":"tencent","releasedAt":"2026-08-20","contextLength":8192,"maxOutput":4096,"input":["text"],"output":["text"],"features":[],"description":"Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided...","inputPerM":0.044,"outputPerM":0.177,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.044,"outputPerM":0.177,"cacheReadPerM":null}]},{"id":"tencent/hy-mt2-30b-a3b","name":"Hy-MT2-30B-A3B","vendor":"tencent","releasedAt":"2026-08-20","contextLength":8192,"maxOutput":4096,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Hy-MT2-30B-A3B is Tencent's flagship translation model in the Hy-MT2 family. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and...","inputPerM":0.074,"outputPerM":0.295,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.074,"outputPerM":0.295,"cacheReadPerM":null}]},{"id":"tencent/hy-mt2-7b","name":"Hy-MT2-7B","vendor":"tencent","releasedAt":"2026-08-19","contextLength":8192,"maxOutput":4096,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Hy-MT2-7B is a 7B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation.","inputPerM":0.074,"outputPerM":0.295,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.074,"outputPerM":0.295,"cacheReadPerM":null}]},{"id":"thedrummer/cydonia-24b-v4.1","name":"Cydonia 24B V4.1","vendor":"thedrummer","releasedAt":"2025-09-27","contextLength":131072,"maxOutput":117964,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.","inputPerM":0.3,"outputPerM":0.5,"cacheReadPerM":0.15,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.3,"outputPerM":0.5,"cacheReadPerM":0.15}]},{"id":"thedrummer/skyfall-36b-v2","name":"Skyfall 36B V2","vendor":"thedrummer","releasedAt":"2025-03-10","contextLength":32768,"maxOutput":29491,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.","inputPerM":0.55,"outputPerM":0.8,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.55,"outputPerM":0.8,"cacheReadPerM":0.25}]},{"id":"thedrummer/unslopnemo-12b","name":"UnslopNemo 12B","vendor":"thedrummer","releasedAt":"2024-11-08","contextLength":1024000,"maxOutput":819200,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","inputPerM":0.4,"outputPerM":0.4,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.4,"outputPerM":0.4,"cacheReadPerM":null}]},{"id":"thinkingmachines/inkling","name":"Inkling","vendor":"thinkingmachines","releasedAt":"2026-07-17","contextLength":524288,"maxOutput":471859,"input":["text","image","audio"],"output":["text"],"features":["reasoning","tools"],"description":"Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems,...","inputPerM":1,"outputPerM":4.05,"cacheReadPerM":0.17,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":4.05,"cacheReadPerM":0.17}]},{"id":"thinkingmachines/inkling-small","name":"Inkling Small","vendor":"thinkingmachines","releasedAt":"2026-07-30","contextLength":524288,"maxOutput":262144,"input":["text","image","audio"],"output":["text"],"features":["reasoning","tools"],"description":"Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...","inputPerM":0.45,"outputPerM":1.2,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.45,"outputPerM":1.2,"cacheReadPerM":0.1}]},{"id":"typesafe/jev-router","name":"Jev Router","vendor":"typesafe","releasedAt":"2026-09-25","contextLength":1000000,"maxOutput":65536,"input":["audio","file","image","text","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Jev Router picks the best model and reasoning effort for each request, balancing quality, speed, and cost. It runs on [Jev](https://openrouter.ai/~typesafe/jev-latest), TypeSafe's first System One model, and adapts as your...","inputPerM":null,"outputPerM":null,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":null,"outputPerM":null,"cacheReadPerM":null}]},{"id":"unbiased/pareto-26.10","name":"Pareto 26.10","vendor":"unbiased","releasedAt":"2026-10-15","contextLength":1048576,"maxOutput":131072,"input":["text","image"],"output":["text"],"features":["tools"],"description":"Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.","inputPerM":0.8,"outputPerM":3.2,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.8,"outputPerM":3.2,"cacheReadPerM":0.03}]},{"id":"unbiased/pareto-26.9","name":"Pareto 26.9","vendor":"unbiased","releasedAt":"2026-10-13","contextLength":262144,"maxOutput":131072,"input":["text","image"],"output":["text"],"features":["tools"],"description":"Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.","inputPerM":2.5,"outputPerM":7.5,"cacheReadPerM":0.25,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.5,"outputPerM":7.5,"cacheReadPerM":0.25}]},{"id":"undi95/remm-slerp-l2-13b","name":"ReMM SLERP 13B","vendor":"undi95","releasedAt":"2023-07-22","contextLength":6144,"maxOutput":5529,"input":["text"],"output":["text"],"features":["structured_outputs"],"description":"A recreation trial of the original MythoMax-L2-B13 but with updated models. #merge","inputPerM":0.35,"outputPerM":0.65,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.35,"outputPerM":0.65,"cacheReadPerM":null}]},{"id":"upstage/solar-mini4","name":"Solar Mini 4","vendor":"upstage","releasedAt":"2026-09-23","contextLength":524288,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Solar Mini 4 is Upstage's compact, cost-efficient language model, a 35B-parameter mixture-of-experts with 3B active parameters and a 524K context window. It is built for agentic use cases where response...","inputPerM":0.05,"outputPerM":0.2,"cacheReadPerM":0.005,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.05,"outputPerM":0.2,"cacheReadPerM":0.005}]},{"id":"upstage/solar-pro-3","name":"Solar Pro 3","vendor":"upstage","releasedAt":"2026-01-27","contextLength":131072,"maxOutput":117964,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Solar Pro 3 is Upstage's powerful Mixture-of-Experts (MoE) language model. With 102B total parameters and 12B active parameters per forward pass, it delivers exceptional performance while maintaining computational efficiency. Optimized...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015,"firstSeen":"2026-10-28","lastSeen":"2026-10-29","gone":"2026-10-29","history":[{"from":"2026-10-28","to":"2026-10-29","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.015}]},{"id":"upstage/solar-pro4","name":"Solar Pro 4","vendor":"upstage","releasedAt":"2026-08-10","contextLength":524288,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Solar Pro 4 is Upstage's cost-efficient large language model, featuring a 524K context window. It is built for long-horizon tasks and agentic workflows, with strong capabilities in office productivity, document-intensive...","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.6,"cacheReadPerM":0.03}]},{"id":"writer/palmyra-x5","name":"Palmyra X5","vendor":"writer","releasedAt":"2026-01-21","contextLength":1040000,"maxOutput":8192,"input":["text"],"output":["text"],"features":[],"description":"Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise. It delivers industry-leading speed and efficiency on context windows up to 1 million...","inputPerM":0.6,"outputPerM":6,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":6,"cacheReadPerM":null}]},{"id":"x-ai/grok-4.20","name":"Grok 4.20","vendor":"x-ai","releasedAt":"2026-03-31","contextLength":2000000,"maxOutput":1800000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Grok 4.20 is a reasoning model from SpaceXAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherance, delivering...","inputPerM":1.25,"outputPerM":2.5,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":2.5,"cacheReadPerM":0.2}]},{"id":"x-ai/grok-4.20-multi-agent","name":"Grok 4.20 Multi-Agent","vendor":"x-ai","releasedAt":"2026-03-31","contextLength":2000000,"maxOutput":1800000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs"],"description":"Grok 4.20 Multi-Agent is a variant of SpaceXAI’s Grok 4.20 designed for collaborative, agent-based workflows. Multiple agents operate in parallel to conduct deep research, coordinate tool use, and synthesize information...","inputPerM":1.25,"outputPerM":2.5,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":2.5,"cacheReadPerM":0.2}]},{"id":"x-ai/grok-4.3","name":"Grok 4.3","vendor":"x-ai","releasedAt":"2026-04-30","contextLength":1000000,"maxOutput":900000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Grok 4.3 is a reasoning model from SpaceXAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual...","inputPerM":1.25,"outputPerM":2.5,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.25,"outputPerM":2.5,"cacheReadPerM":0.2}]},{"id":"x-ai/grok-4.5","name":"Grok 4.5","vendor":"x-ai","releasedAt":"2026-07-08","contextLength":500000,"maxOutput":450000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Grok 4.5 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM.","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.3,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.3}]},{"id":"x-ai/grok-4.6","name":"Grok 4.6","vendor":"x-ai","releasedAt":"2026-08-12","contextLength":500000,"maxOutput":450000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Grok 4.6 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7).","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.5}]},{"id":"x-ai/grok-4.7","name":"Grok 4.7","vendor":"x-ai","releasedAt":"2026-09-21","contextLength":500000,"maxOutput":450000,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.5,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2,"outputPerM":6,"cacheReadPerM":0.5}]},{"id":"x-ai/grok-build-0.1","name":"Grok Build 0.1","vendor":"x-ai","releasedAt":"2026-05-20","contextLength":256000,"maxOutput":230400,"input":["text","image","file"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Grok Build 0.1 is SpaceXAI’s fast coding model trained specifically for agentic software engineering workflows. It supports text and image inputs with text output, and is optimized for interactive coding...","inputPerM":1,"outputPerM":2,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":2,"cacheReadPerM":0.2}]},{"id":"xiaomi/mimo-v2.5","name":"MiMo-V2.5","vendor":"xiaomi","releasedAt":"2026-04-22","contextLength":1000000,"maxOutput":65536,"input":["text","audio","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.08,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.4,"outputPerM":2,"cacheReadPerM":0.08}]},{"id":"xiaomi/mimo-v2.5-pro","name":"MiMo-V2.5-Pro","vendor":"xiaomi","releasedAt":"2026-04-22","contextLength":1050000,"maxOutput":235929,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as ClawEval, GDPVal, and SWE-bench Pro....","inputPerM":0.48,"outputPerM":1.8,"cacheReadPerM":0.096,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.48,"outputPerM":1.8,"cacheReadPerM":0.096}]},{"id":"xiaomi/mimo-v2.6-flash","name":"MiMo-V2.6-Flash","vendor":"xiaomi","releasedAt":"2026-09-21","contextLength":1050000,"maxOutput":131072,"input":["text","image","video","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for...","inputPerM":0.14,"outputPerM":0.28,"cacheReadPerM":0.0028,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.14,"outputPerM":0.28,"cacheReadPerM":0.0028}]},{"id":"xiaomi/mimo-v2.6-pro","name":"MiMo-V2.6-Pro","vendor":"xiaomi","releasedAt":"2026-09-21","contextLength":1050000,"maxOutput":131072,"input":["text","image","video","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiMo-V2.6-Pro is the flagship foundation model developed by Xiaomi. Built at a scale of over 1T parameters, it is designed to push the ceiling of capability for the most demanding...","inputPerM":0.435,"outputPerM":0.87,"cacheReadPerM":0.0036,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.435,"outputPerM":0.87,"cacheReadPerM":0.0036}]},{"id":"xiaomi/mimo-v2.6-pro-ultraspeed","name":"MiMo-V2.6-Pro-UltraSpeed","vendor":"xiaomi","releasedAt":"2026-09-21","contextLength":1048576,"maxOutput":131072,"input":["text","image","video","audio"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"MiMo-V2.6-Pro-UltraSpeed is the fast speed edition of Xiaomi's flagship foundation model, MiMo-V2.6-Pro. Built from the same 1T MiMo-V2.6-Pro checkpoint, it matches the original model in quality while delivering roughly 10x...","inputPerM":4.35,"outputPerM":8.7,"cacheReadPerM":0.036,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":4.35,"outputPerM":8.7,"cacheReadPerM":0.036}]},{"id":"z-ai/glm-4.5","name":"GLM 4.5","vendor":"z-ai","releasedAt":"2025-07-25","contextLength":131072,"maxOutput":98304,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications. It leverages a Mixture-of-Experts (MoE) architecture and supports a context length of up to 128k tokens. GLM-4.5 delivers significantly...","inputPerM":0.6,"outputPerM":2.2,"cacheReadPerM":0.11,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":2.2,"cacheReadPerM":0.11}]},{"id":"z-ai/glm-4.5-air","name":"GLM 4.5 Air","vendor":"z-ai","releasedAt":"2025-07-25","contextLength":131072,"maxOutput":98304,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications. Like GLM-4.5, it adopts the Mixture-of-Experts (MoE) architecture but with a more compact parameter...","inputPerM":0.13,"outputPerM":0.85,"cacheReadPerM":0.025,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.13,"outputPerM":0.85,"cacheReadPerM":0.025}]},{"id":"z-ai/glm-4.5v","name":"GLM 4.5V","vendor":"z-ai","releasedAt":"2025-08-11","contextLength":65536,"maxOutput":16384,"input":["text","image"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-4.5V is a vision-language foundation model for multimodal agent applications. Built on a Mixture-of-Experts (MoE) architecture with 106B parameters and 12B activated parameters, it achieves state-of-the-art results in video understanding,...","inputPerM":0.6,"outputPerM":1.8,"cacheReadPerM":0.11,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.6,"outputPerM":1.8,"cacheReadPerM":0.11}]},{"id":"z-ai/glm-4.6","name":"GLM 4.6","vendor":"z-ai","releasedAt":"2025-09-30","contextLength":204800,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...","inputPerM":0.5,"outputPerM":2,"cacheReadPerM":0.1,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.55,"outputPerM":2.2,"cacheReadPerM":0.11},{"from":"2026-10-29","to":"2026-11-06","inputPerM":0.5,"outputPerM":2,"cacheReadPerM":0.1}]},{"id":"z-ai/glm-4.6v","name":"GLM 4.6V","vendor":"z-ai","releasedAt":"2025-12-08","contextLength":131072,"maxOutput":32768,"input":["image","text","video"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed media. It supports up to 128K tokens, processes complex page layouts...","inputPerM":0.3,"outputPerM":0.9,"cacheReadPerM":0.055,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-29","inputPerM":0.3,"outputPerM":0.9,"cacheReadPerM":0.05},{"from":"2026-10-30","to":"2026-11-06","inputPerM":0.3,"outputPerM":0.9,"cacheReadPerM":0.055}]},{"id":"z-ai/glm-4.7","name":"GLM 4.7","vendor":"z-ai","releasedAt":"2025-12-22","contextLength":204800,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution. It demonstrates significant improvements in executing complex agent tasks while...","inputPerM":0.6,"outputPerM":2.2,"cacheReadPerM":0.11,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-31","inputPerM":0.54,"outputPerM":1.98,"cacheReadPerM":0.099},{"from":"2026-10-31","to":"2026-11-06","inputPerM":0.6,"outputPerM":2.2,"cacheReadPerM":0.11}]},{"id":"z-ai/glm-4.7-flash","name":"GLM 4.7 Flash","vendor":"z-ai","releasedAt":"2026-01-19","contextLength":200000,"maxOutput":117964,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency. It is further optimized for agentic coding use cases, strengthening coding capabilities, long-horizon task planning,...","inputPerM":0.0605,"outputPerM":0.4,"cacheReadPerM":null,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":0.0605,"outputPerM":0.4,"cacheReadPerM":null}]},{"id":"z-ai/glm-5","name":"GLM 5","vendor":"z-ai","releasedAt":"2026-02-11","contextLength":204800,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows. Built for expert developers, it delivers production-grade performance on large-scale programming tasks, rivaling leading...","inputPerM":1,"outputPerM":3.2,"cacheReadPerM":0.2,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1,"outputPerM":3.2,"cacheReadPerM":0.2}]},{"id":"z-ai/glm-5.1","name":"GLM 5.1","vendor":"z-ai","releasedAt":"2026-04-07","contextLength":204800,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks. Unlike previous models built around minute-level interactions, GLM-5.1 can work independently and continuously on...","inputPerM":1.4,"outputPerM":4.4,"cacheReadPerM":0.26,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.4,"outputPerM":4.4,"cacheReadPerM":0.26}]},{"id":"z-ai/glm-5.2","name":"GLM 5.2","vendor":"z-ai","releasedAt":"2026-06-16","contextLength":1048576,"maxOutput":943718,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...","inputPerM":0.006,"outputPerM":6.8,"cacheReadPerM":0.005,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.014,"outputPerM":7,"cacheReadPerM":0.013},{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.0106,"outputPerM":7,"cacheReadPerM":0.0096},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.006,"outputPerM":5.4,"cacheReadPerM":0.005},{"from":"2026-10-29","to":"2026-10-31","inputPerM":0.006,"outputPerM":7,"cacheReadPerM":0.005},{"from":"2026-10-31","to":"2026-11-06","inputPerM":0.006,"outputPerM":6.8,"cacheReadPerM":0.005}]},{"id":"z-ai/glm-5.3","name":"GLM 5.3","vendor":"z-ai","releasedAt":"2026-08-18","contextLength":1048576,"maxOutput":943718,"input":["text"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...","inputPerM":0.0699,"outputPerM":1.4,"cacheReadPerM":0.0689,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-29","inputPerM":0.0699,"outputPerM":1.4,"cacheReadPerM":0.0689},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.08,"outputPerM":3.6,"cacheReadPerM":0.076},{"from":"2026-10-31","to":"2026-11-01","inputPerM":0.0699,"outputPerM":1.4,"cacheReadPerM":0.0689},{"from":"2026-11-01","to":"2026-11-04","inputPerM":0.08,"outputPerM":3.6,"cacheReadPerM":0.076},{"from":"2026-11-04","to":"2026-11-06","inputPerM":0.0699,"outputPerM":1.4,"cacheReadPerM":0.0689}]},{"id":"z-ai/glm-5.3-flash","name":"GLM 5.3 Flash","vendor":"z-ai","releasedAt":"2026-08-26","contextLength":1048576,"maxOutput":943717,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...","inputPerM":0.15,"outputPerM":0.5,"cacheReadPerM":0.03,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-10-28","inputPerM":0.15,"outputPerM":0.5,"cacheReadPerM":0.03},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.0379,"outputPerM":0.0798,"cacheReadPerM":0.03},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.0457,"outputPerM":0.0742,"cacheReadPerM":0.03},{"from":"2026-10-29","to":"2026-10-29","inputPerM":0.15,"outputPerM":0.5,"cacheReadPerM":0.03},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.0481,"outputPerM":0.146,"cacheReadPerM":0.0399},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.0457,"outputPerM":0.0701,"cacheReadPerM":0.03},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.15,"outputPerM":0.5,"cacheReadPerM":0.03},{"from":"2026-10-30","to":"2026-10-30","inputPerM":0.018,"outputPerM":0.1402,"cacheReadPerM":0.018},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.0216,"outputPerM":0.1238,"cacheReadPerM":0.0216},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.0163,"outputPerM":0.1402,"cacheReadPerM":0.0163},{"from":"2026-10-31","to":"2026-10-31","inputPerM":0.0163,"outputPerM":0.1479,"cacheReadPerM":0.0163},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.0143,"outputPerM":0.0184,"cacheReadPerM":0.013},{"from":"2026-11-01","to":"2026-11-01","inputPerM":0.0163,"outputPerM":0.1821,"cacheReadPerM":0.0163},{"from":"2026-11-01","to":"2026-11-03","inputPerM":0.0285,"outputPerM":0.0923,"cacheReadPerM":0.0285},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.0313,"outputPerM":0.0846,"cacheReadPerM":0.03},{"from":"2026-11-03","to":"2026-11-03","inputPerM":0.033,"outputPerM":0.242,"cacheReadPerM":0.022},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.0313,"outputPerM":0.0846,"cacheReadPerM":0.03},{"from":"2026-11-04","to":"2026-11-04","inputPerM":0.0269,"outputPerM":0.0996,"cacheReadPerM":0.0269},{"from":"2026-11-04","to":"2026-11-06","inputPerM":0.15,"outputPerM":0.5,"cacheReadPerM":0.03}]},{"id":"z-ai/glm-5.3-flashx","name":"GLM 5.3 FlashX","vendor":"z-ai","releasedAt":"2026-09-18","contextLength":1048576,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s. Built on the same hybrid sparse and linear attention architecture...","inputPerM":0.37,"outputPerM":1.25,"cacheReadPerM":0.075,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-02","inputPerM":0.37,"outputPerM":1.25,"cacheReadPerM":0.09},{"from":"2026-11-02","to":"2026-11-06","inputPerM":0.37,"outputPerM":1.25,"cacheReadPerM":0.075}]},{"id":"z-ai/glm-5.3-flashx-legal","name":"GLM 5.3 FlashX Legal","vendor":"z-ai","releasedAt":"2026-11-02","contextLength":1048576,"maxOutput":131072,"input":["text","image","video"],"output":["text"],"features":["reasoning","structured_outputs","tools"],"description":"GLM-5.3-FlashX-Legal (GLM-Legal) is Z.ai's legal-domain variant of GLM-5.3-FlashX, a native multimodal model that accepts text, image, and video input and produces text. It adds legal-domain mid-training and post-training with reinforcement...","inputPerM":0.37,"outputPerM":1.25,"cacheReadPerM":0.075,"firstSeen":"2026-11-02","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-11-02","to":"2026-11-02","inputPerM":0.37,"outputPerM":1.25,"cacheReadPerM":0.09},{"from":"2026-11-02","to":"2026-11-06","inputPerM":0.37,"outputPerM":1.25,"cacheReadPerM":0.075}]},{"id":"z-ai/glm-5.3-prime","name":"GLM 5.3 Prime","vendor":"z-ai","releasedAt":"2026-09-23","contextLength":1000000,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token...","inputPerM":2.8,"outputPerM":8.8,"cacheReadPerM":0.56,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":2.8,"outputPerM":8.8,"cacheReadPerM":0.56}]},{"id":"z-ai/glm-5-turbo","name":"GLM 5 Turbo","vendor":"z-ai","releasedAt":"2026-03-15","contextLength":202752,"maxOutput":131072,"input":["text"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios. It is deeply optimized for real-world agent workflows...","inputPerM":1.2,"outputPerM":4,"cacheReadPerM":0.24,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.2,"outputPerM":4,"cacheReadPerM":0.24}]},{"id":"z-ai/glm-5v-turbo","name":"GLM 5V Turbo","vendor":"z-ai","releasedAt":"2026-04-01","contextLength":202752,"maxOutput":131072,"input":["image","text","video"],"output":["text"],"features":["reasoning","tools"],"description":"GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks. It natively handles image, video, and text inputs, excels at long-horizon planning, complex coding,...","inputPerM":1.2,"outputPerM":4,"cacheReadPerM":0.24,"firstSeen":"2026-10-28","lastSeen":"2026-11-06","gone":null,"history":[{"from":"2026-10-28","to":"2026-11-06","inputPerM":1.2,"outputPerM":4,"cacheReadPerM":0.24}]}],"priceKnowledge":{"day":"2026-10-27","tools":300,"exact":0.1,"close":0.35,"freshDays":60,"raw":"content/research/model-price-knowledge-2026-10-27.json","models":[{"model":"openai/gpt-6-luna","asked":300,"failed":0,"right":26.7,"near":14,"wrong":22.3,"unknown":37,"freePlanRight":78.7,"cost":0.0382},{"model":"anthropic/claude-haiku-5.5","asked":300,"failed":0,"right":19.7,"near":10.3,"wrong":17.7,"unknown":52.3,"freePlanRight":79.3,"cost":0.0343},{"model":"google/gemini-3.5-flash-lite","asked":300,"failed":0,"right":17.3,"near":16,"wrong":25.7,"unknown":41,"freePlanRight":79.5,"cost":0.0551},{"model":"deepseek/deepseek-v4.1-flash","asked":296,"failed":4,"right":20.6,"near":14.5,"wrong":28,"unknown":36.8,"freePlanRight":78.9,"cost":0.2145},{"model":"meta-llama/llama-4-maverick","asked":300,"failed":0,"right":13.3,"near":12.3,"wrong":43,"unknown":31.3,"freePlanRight":74.5,"cost":0.0149},{"model":"mistralai/mistral-small-2603","asked":270,"failed":30,"right":5.9,"near":11.1,"wrong":24.1,"unknown":58.9,"freePlanRight":69.2,"cost":0.0595},{"model":"x-ai/grok-4.3","asked":300,"failed":0,"right":1.7,"near":2,"wrong":0.7,"unknown":95.7,"freePlanRight":89.7,"cost":0.4455},{"model":"qwen/qwen3.8-flash","asked":298,"failed":2,"right":12.8,"near":9.1,"wrong":19.5,"unknown":58.7,"freePlanRight":80.9,"cost":0.2249},{"model":"z-ai/glm-5.3-flash","asked":300,"failed":0,"right":13.3,"near":9,"wrong":14.7,"unknown":63,"freePlanRight":78.7,"cost":0.0079},{"model":"amazon/nova-2-lite-v1","asked":300,"failed":0,"right":11.3,"near":21.3,"wrong":53.7,"unknown":13.7,"freePlanRight":73.4,"cost":0.5166},{"model":"moonshotai/kimi-k2.5","asked":300,"failed":0,"right":8.3,"near":6.7,"wrong":9.3,"unknown":75.7,"freePlanRight":79.1,"cost":1.4487}]}}