{"object":"list","data":[{"id":"alibaba/qwen3.8-2.4t-a95b","object":"model","created":1755815280,"released":1785715200,"owned_by":"alibaba","name":"Qwen3.8 2.4T A95B","description":"Open-weights release of the Qwen3.8 flagship (2.4T MoE, ~95B active); thinking always on with reasoning_effort low/medium/xhigh. The hosted Qwen 3.8 Max (vision, non-thinking, 1M default context) is Alibaba-only.","context_window":262144,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000021","output":"0.0000063"}},{"id":"deepseek/deepseek-v4-flash-vision-exp","object":"model","created":1755815280,"released":1787270400,"owned_by":"deepseek","name":"DeepSeek V4 Flash Vision Exp","description":"DeepSeek-V4-Flash-Vision-Exp is an experimental multimodal model that combines the agentic reasoning, coding, and world knowledge capabilities of DeepSeek-V4-Flash with advanced visual understanding. It delivers major gains on vision-dependent agent workflows while performing strongly across leading coding, repository, data science, automation, and multimodal benchmarks.","context_window":1000000,"max_tokens":384000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"temperature":true,"pricing":{"input_cache_read":"0.00000000735","input":"0.000000231","output":"0.000000693"}},{"id":"alibaba/qwen-3-32b","object":"model","created":1755815280,"released":1745798400,"owned_by":"alibaba","name":"Qwen 3 32B","description":"Qwen3-32B is a world-class model with comparable quality to DeepSeek R1 while outperforming GPT-4.1 and Claude Sonnet 3.7. It excels in code-gen, tool-calling, and advanced reasoning, making it an exceptional model for a wide range of production use cases.","context_window":128000,"max_tokens":8192,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":38912}],"temperature":true,"pricing":{"input":"0.000000168","output":"0.000000672"}},{"id":"alibaba/qwen3.7-flash","object":"model","created":1755815280,"released":1785196800,"owned_by":"alibaba","name":"Qwen 3.7 Flash","description":"The Qwen3.7 native vision-language Flash model series delivers a comprehensive upgrade over 3.6-Flash in multimodal understanding and agent execution. This model particularly excels in enhanced multimodal foundations with stronger universal object recognition, further improved real-world perception and spatial intelligence, significantly upgraded multimodal agent capabilities for Search Agent and CI Agent scenarios with more stable end-to-end task execution, as well as optimized multimodal coding for a smoother vibe coding experience.","context_window":991000,"max_tokens":64000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision"],"temperature":true,"pricing":{"input_cache_read":"0.0000000063","input_cache_write":"0.0000000399","input":"0.0000000315","output":"0.0000001365"}},{"id":"amazon/nova-pro","object":"model","created":1755815280,"released":1733184000,"owned_by":"amazon","name":"Nova Pro","description":"A highly capable multimodal model with the best combination of accuracy, speed, and cost for a wide range of tasks.","context_window":300000,"max_tokens":8192,"type":"language","tags":["file-input","tool-use","vision"],"temperature":true,"pricing":{"input":"0.00000084","output":"0.00000336"}},{"id":"anthropic/claude-3-haiku","object":"model","created":1755815280,"released":1710288000,"owned_by":"anthropic","name":"Claude 3 Haiku","description":"Claude 3 Haiku is Anthropic's fastest, most compact model for near-instant responsiveness. It answers simple queries and requests with speed. Customers will be able to build seamless AI experiences that mimic human interactions. Claude 3 Haiku can process images and return text outputs, and features a 200K context window.","context_window":200000,"max_tokens":4096,"type":"language","tags":["tool-use","vision","explicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input_cache_write":"0.000000315","input":"0.0000002625","output":"0.0000013125"}},{"id":"bytedance/seed-1.6","object":"model","created":1755815280,"released":1756684800,"owned_by":"bytedance","name":"Seed 1.6","description":"ByteDance's new multimodal deep-thinking model, supporting both text and visual inputs with enhanced reasoning capabilities.","context_window":256000,"max_tokens":32000,"type":"language","tags":["reasoning","tool-use","vision"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.0000000525","input":"0.0000002625","output":"0.0000021"}},{"id":"google/gemini-3-flash","object":"model","created":1755815280,"released":1765929600,"owned_by":"google","name":"Gemini 3 Flash","description":"Google's most intelligent model built for speed, combining frontier intelligence with superior search and grounding.","context_window":1000000,"max_tokens":65000,"type":"language","tags":["reasoning","file-input","vision","tool-use","web-search","implicit-caching"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000000525","input":"0.000000525","output":"0.00000315","web_search":"14.7"}},{"id":"google/gemini-3.1-flash-image","object":"model","created":1755815280,"released":1779926400,"owned_by":"google","name":"Gemini 3.1 Flash Image (Nano Banana 2)","description":"Gemini 3.1 Flash Image is optimized for image understanding and generation and offers a balance of price and performance.","context_window":131072,"max_tokens":32768,"type":"language","tags":["image-generation","implicit-caching","reasoning","vision","web-search"],"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000000525","input":"0.000000525","output":"0.00000315","web_search":"14.7"}},{"id":"openai/o1","object":"model","created":1755815280,"released":1733356800,"owned_by":"openai","name":"o1","description":"o1 is OpenAI's flagship reasoning model, designed for complex problems that require deep thinking. It provides strong reasoning capabilities with improved accuracy for complex multi-step tasks.","context_window":200000,"max_tokens":100000,"type":"language","tags":["file-input","reasoning","tool-use","vision","implicit-caching"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.000007875","input":"0.00001575","output":"0.000063"}},{"id":"alibaba/qwen3-coder-30b-a3b","object":"model","created":1755815280,"released":1753920000,"owned_by":"alibaba","name":"Qwen 3 Coder 30B A3B Instruct","description":"Efficient coding specialist balancing performance with cost-effectiveness for daily development tasks while maintaining strong tool integration capabilities.","context_window":262144,"max_tokens":8192,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.00000063"}},{"id":"alibaba/qwen3-coder-plus","object":"model","created":1755815280,"released":1753228800,"owned_by":"alibaba","name":"Qwen3 Coder Plus","description":"Powered by Qwen3 this is a powerful Coding Agent that excels in tool calling and environment interaction to achieve autonomous programming. It combines outstanding coding proficiency with versatile general-purpose abilities.","context_window":1000000,"max_tokens":65536,"type":"language","tags":["implicit-caching","tool-use"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.00000105","output":"0.00000525"}},{"id":"alibaba/qwen3-max","object":"model","created":1755815280,"released":1758585600,"owned_by":"alibaba","name":"Qwen3 Max","description":"The Qwen 3 series Max model has undergone specialized upgrades in agent programming and tool invocation compared to the preview version. The officially released model this time has achieved state-of-the-art (SOTA) performance in its field and is better suited to meet the demands of agents operating in more complex scenarios.","context_window":262144,"max_tokens":32768,"type":"language","tags":["tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000252","input":"0.00000126","output":"0.0000063"}},{"id":"google/gemini-3.7-flash","object":"model","created":1755815280,"released":1786579200,"owned_by":"google","name":"Gemini 3.7 Flash","description":"Gemini 3.7 Flash is the high-efficiency, cost-effective powerhouse of the Gemini 3 family. It delivers Pro-level agentic capabilities, major leaps in code generation and terminal execution. 3.7 Flash serves as the primary agentic workhorse in the Gemini 3 family, bridging the gap between deep-reasoning Pro models and high-throughput Flash-Lite models while delivering high token efficiency and multi-step multimodal processing.","context_window":1000000,"max_tokens":65536,"type":"language","tags":["reasoning","file-input","vision","tool-use","web-search","implicit-caching","video-input"],"temperature":true,"pricing":{"input_cache_read":"0.00000007875","input":"0.0000007875","output":"0.0000039375"}},{"id":"kwaipilot/kat-coder-pro-v1","object":"model","created":1755815280,"released":1762646400,"owned_by":"kwaipilot","name":"KAT-Coder-Pro V1","description":"KAT-Coder-Pro V1 is KwaiKAT's most advanced agentic coding model in the KwaiKAT series. Designed specifically for agentic coding tasks, it excels in real-world software engineering scenarios, achieving a remarkable 73.4% solve rate on the SWE-Bench Verified benchmark. KAT-Coder-Pro V1 delivers top-tier coding performance and has been rigorously tested by thousands of in-house engineers. The model has been optimized for tool-use capability, multi-turn interaction, instruction following, generalization and comprehensive capabilities through a multi-stage training process, including mid-training, supervised fine-tuning (SFT), reinforcement fine-tuning (RFT), and scalable agentic RL.","context_window":256000,"max_tokens":32000,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input_cache_read":"0.000000063","input":"0.000000315","output":"0.00000126"}},{"id":"google/gemini-3.1-flash-lite","object":"model","created":1755815280,"released":1778112000,"owned_by":"google","name":"Gemini 3.1 Flash Lite","description":"Gemini 3.1 Flash Lite outperforms 2.5 Flash Lite on overall quality and lands close to 2.5 Flash performance across key capability areas. It is a workhorse model for high-volume use cases, with improvements across audio input/ASR, RAG snippet ranking, translation, data extraction, and code completion.","context_window":1000000,"max_tokens":65000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","web-search"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.0000002625","output":"0.000001575","web_search":"14.7"}},{"id":"google/gemini-3.5-flash-lite","object":"model","created":1755815280,"released":1784592000,"owned_by":"google","name":"Gemini 3.5 Flash Lite","description":"Gemini 3.5 Flash Lite features upgraded agentic capabilities, making the model ideal for subagents in complex workflows.","context_window":1000000,"max_tokens":65000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","web-search","video-input"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.000000315","output":"0.000002625","web_search":"14.7"}},{"id":"alibaba/qwen3-max-preview","object":"model","created":1755815280,"released":1757030400,"owned_by":"alibaba","name":"Qwen3 Max Preview","description":"Qwen3-Max-Preview shows substantial gains over the 2.5 series in overall capability, with significant enhancements in Chinese-English text understanding, complex instruction following, handling of subjective open-ended tasks, multilingual ability, and tool invocation; model knowledge hallucinations are reduced.","context_window":262144,"max_tokens":32768,"type":"language","tags":["tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000252","input":"0.00000126","output":"0.0000063"}},{"id":"alibaba/qwen3-235b-a22b-thinking","object":"model","created":1755815280,"released":1758585600,"owned_by":"alibaba","name":"Qwen3 VL 235B A22B Thinking","description":"Qwen3 series VL models feature significantly enhanced multimodal reasoning capabilities, with a particular focus on optimizing the model for STEM and mathematical reasoning. Visual perception and recognition abilities have been comprehensively improved, and OCR capabilities have undergone a major upgrade.","context_window":131072,"max_tokens":32768,"type":"language","tags":["vision","reasoning","tool-use","file-input"],"reasoning_options":[{"type":"budget_tokens","min":1,"max":81920}],"temperature":true,"pricing":{"input":"0.00000042","output":"0.0000042"}},{"id":"alibaba/qwen3-coder","object":"model","created":1755815280,"released":1753142400,"owned_by":"alibaba","name":"Qwen3 Coder 480B A35B Instruct","description":"Qwen3-Coder-480B-A35B-Instruct is a cutting-edge open coding model from Qwen, matching Claude Sonnet’s performance in agentic programming, browser automation, and core development tasks.","context_window":262144,"max_tokens":65536,"type":"language","tags":["tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000315","input":"0.000001575","output":"0.000007875"}},{"id":"alibaba/qwen3.6-27b","object":"model","created":1755815280,"released":1776816000,"owned_by":"alibaba","name":"Qwen 3.6 27B","description":"The Qwen3.6 35B-A3B native vision-language model is built on a hybrid architecture that integrates linear attention mechanisms with a sparse mixture-of-experts framework, achieving higher inference efficiency. Compared with the 3.5-35B-A3B, this model demonstrates significantly improved agentic coding capabilities, mathematical and code reasoning abilities, spatial intelligence, as well as object localization and object detection performance.","context_window":256000,"max_tokens":256000,"type":"language","tags":["reasoning","tool-use","file-input","vision"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":131072}],"temperature":true,"pricing":{"input":"0.00000063","output":"0.00000378"}},{"id":"openai/gpt-4.1-fast","object":"model","created":1755815280,"released":1744588800,"owned_by":"openai","name":"GPT-4.1 (Fast)","description":"GPT 4.1 is OpenAI's flagship model for complex tasks. It is well suited for problem solving across domains.","context_window":1047576,"max_tokens":32768,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.00000091875","input":"0.000003675","output":"0.0000147","web_search":"10.5"}},{"id":"openai/gpt-4.1-nano","object":"model","created":1755815280,"released":1744588800,"owned_by":"openai","name":"GPT-4.1 nano","description":"GPT-4.1 nano is the fastest, most cost-effective GPT 4.1 model.","context_window":1047576,"max_tokens":32768,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.00000002625","input":"0.000000105","output":"0.00000042","web_search":"10.5"}},{"id":"alibaba/qwen-3-14b","object":"model","created":1755815280,"released":1745798400,"owned_by":"alibaba","name":"Qwen3-14B","description":"Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support","context_window":40960,"max_tokens":16384,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.000000126","output":"0.000000252"}},{"id":"alibaba/qwen3-coder-next","object":"model","created":1755815280,"released":1753142400,"owned_by":"alibaba","name":"Qwen3 Coder Next","description":"Qwen3-Coder-Next is an open-weight language model built specifically for coding, with strong performance on large-scale software engineering and agentic coding benchmarks. It uses a hybrid Mixture-of-Experts architecture to offer high capability at relatively modest active parameter counts, improving efficiency for real-world deployments. The model is trained on diverse code and natural language data so it can handle tasks like code generation, refactoring, debugging, repository-level reasoning, and technical explanation across multiple programming languages. It is also optimized for tool use and function calling, making it suitable as the core of coding agents that interact with shells, editors, issue trackers, and other developer tools.","context_window":256000,"max_tokens":256000,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.000000525","output":"0.00000126"}},{"id":"alibaba/qwen3-next-80b-a3b-instruct","object":"model","created":1755815280,"released":1757548800,"owned_by":"alibaba","name":"Qwen3 Next 80B A3B Instruct","description":"A new generation of open-source, non-thinking mode model powered by Qwen3. This version demonstrates superior Chinese text understanding, augmented logical reasoning, and enhanced capabilities in text generation tasks over the previous iteration (Qwen3-235B-A22B-Instruct-2507).","context_window":131072,"max_tokens":32768,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.00000126"}},{"id":"alibaba/qwen3-next-80b-a3b-thinking","object":"model","created":1755815280,"released":1757548800,"owned_by":"alibaba","name":"Qwen3 Next 80B A3B Thinking","description":"A new generation of Qwen3-based open-source thinking mode models. This version offers improved instruction following and streamlined summary responses over the previous iteration (Qwen3-235B-A22B-Thinking-2507).","context_window":131072,"max_tokens":32768,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"budget_tokens","min":1}],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.00000126"}},{"id":"alibaba/qwen3-vl-thinking","object":"model","created":1755815280,"released":1758585600,"owned_by":"alibaba","name":"Qwen3 VL 235B A22B Thinking","description":"Qwen3 series VL models feature significantly enhanced multimodal reasoning capabilities, with a particular focus on optimizing the model for STEM and mathematical reasoning. Visual perception and recognition abilities have been comprehensively improved, and OCR capabilities have undergone a major upgrade.","context_window":131072,"max_tokens":32768,"type":"language","tags":["vision","reasoning","tool-use","file-input"],"reasoning_options":[{"type":"budget_tokens","min":1,"max":81920}],"temperature":true,"pricing":{"input":"0.00000042","output":"0.0000042"}},{"id":"alibaba/qwen3.7-max","object":"model","created":1755815280,"released":1779321600,"owned_by":"alibaba","name":"Qwen 3.7 Max","description":"Qwen3.7 is a next‑generation flagship model designed for the agent‑centric era, with its core strengths lying in the breadth and depth of its agent‑level capabilities: it excels at programming, office and productivity tasks, and long‑term autonomous execution.","context_window":991000,"max_tokens":64000,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":262144}],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input_cache_write":"0.00000328125","input":"0.000002625","output":"0.000007875"}},{"id":"alibaba/qwen3.7-plus","object":"model","created":1755815280,"released":1780358400,"owned_by":"alibaba","name":"Qwen 3.7 Plus","description":"Among the Qwen3.7 series, the cost-effective Plus model builds on its robust text capabilities while delivering a comprehensive upgrade to its vision‑language abilities, all while preserving its full‑stack agent‑level intelligence for coding, tool use, and productivity workflows.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":262144}],"temperature":true,"pricing":{"input_cache_read":"0.000000084","input_cache_write":"0.000000525","input":"0.00000042","output":"0.00000168"}},{"id":"anthropic/claude-fable-5","object":"model","created":1755815280,"released":1782864000,"owned_by":"anthropic","name":"Claude Fable 5","description":"Claude Fable 5 is a Mythos-class model with robust safeguards. It can handle long-running, complex, and asynchronous tasks where previous models would have needed more frequent check-ins.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","explicit-caching","file-input","vision"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000105","input_cache_write":"0.000013125","input":"0.0000105","output":"0.0000525","web_search":"10.5"}},{"id":"anthropic/claude-haiku-4.5","object":"model","created":1755815280,"released":1760486400,"owned_by":"anthropic","name":"Claude Haiku 4.5","description":"Claude Haiku 4.5 matches Sonnet 4's performance on coding, computer use, and agent tasks at substantially lower cost and faster speeds. It delivers near-frontier performance and Claude’s unique character at a price point that works for scaled sub-agent deployments, free tier products, and intelligence-sensitive applications with budget constraints.","context_window":200000,"max_tokens":64000,"type":"language","tags":["explicit-caching","file-input","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024}],"temperature":true,"pricing":{"input_cache_read":"0.000000105","input_cache_write":"0.0000013125","input":"0.00000105","output":"0.00000525","web_search":"10.5"}},{"id":"anthropic/claude-opus-4","object":"model","created":1755815280,"released":1747872000,"owned_by":"anthropic","name":"Claude Opus 4","description":"Claude Opus 4 is Anthropic's most powerful model yet and the state-of-the-art coding model. It delivers sustained performance on long-running tasks that require focused effort and thousands of steps, significantly expanding what AI agents can solve. Claude Opus 4 is ideal for powering frontier agent products and features.","context_window":200000,"max_tokens":8192,"type":"language","tags":["file-input","reasoning","tool-use","vision","explicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000001575","input_cache_write":"0.0000196875","input":"0.00001575","output":"0.00007875","web_search":"10.5"}},{"id":"alibaba/qwen3.5-flash","object":"model","created":1755815280,"released":1771891200,"owned_by":"alibaba","name":"Qwen 3.5 Flash","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 3 series, these models deliver a leap forward in performance for both pure text and multimodal tasks, offering fast response times while balancing inference speed and overall performance.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["vision","file-input","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":81920}],"temperature":true,"pricing":{"input_cache_read":"0.00000000105","input_cache_write":"0.00000013125","input":"0.000000105","output":"0.00000042"}},{"id":"anthropic/claude-opus-4.6","object":"model","created":1755815280,"released":1770249600,"owned_by":"anthropic","name":"Claude Opus 4.6","description":"Opus 4.6 is the world’s best model for coding and professional work, built to power agents that take on whole categories of real-world work. It excels across the entire SDLC, breaking through on hard problems, identifying complex bugs, and demonstrating deeper codebase understanding. It also delivers a step-change in knowledge work, with near-production-ready documents, presentations, and spreadsheets on the first pass.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["tool-use","reasoning","vision","file-input","explicit-caching","web-search"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024}],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input_cache_write":"0.0000065625","input":"0.00000525","output":"0.00002625","web_search":"10.5"}},{"id":"anthropic/claude-opus-4.8-fast","object":"model","created":1755815280,"released":1779926400,"owned_by":"anthropic","name":"Claude Opus 4.8 (Fast)","description":"Opus 4.8 is a focused upgrade to Opus 4.7 and is Anthropic's best generally available model for coding, agentic tasks, and enterprise workflows. It builds on the strengths of previous Opus models with stronger performance on complex, multi-step coding tasks. Anthropic recommends using it on long-horizon coding and agentic tasks. It is also stronger on professional work, including document drafting, data analysis, and presentations.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["tool-use","reasoning","vision","file-input","explicit-caching","web-search","fast"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000105","input_cache_write":"0.000013125","input":"0.0000105","output":"0.0000525","web_search":"10.5"}},{"id":"moonshotai/kimi-k2","object":"model","created":1755815280,"released":1752192000,"owned_by":"moonshotai","name":"Kimi K2 Instruct","description":"Kimi K2 is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters. Trained with the Muon optimizer, Kimi K2 achieves exceptional performance across frontier knowledge, reasoning, and coding tasks while being meticulously optimized for agentic capabilities.","context_window":131072,"max_tokens":131072,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.0000005985","output":"0.000002415"}},{"id":"anthropic/claude-sonnet-4.5","object":"model","created":1755815280,"released":1759104000,"owned_by":"anthropic","name":"Claude Sonnet 4.5","description":"Claude Sonnet 4.5 is the newest model in the Sonnet series, offering improvements and updates over Sonnet 4.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["explicit-caching","file-input","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"budget_tokens","min":1024}],"temperature":true,"pricing":{"input_cache_read":"0.000000315","input_cache_write":"0.0000039375","input":"0.00000315","output":"0.00001575","web_search":"10.5"}},{"id":"anthropic/claude-sonnet-4.6","object":"model","created":1755815280,"released":1771286400,"owned_by":"anthropic","name":"Claude Sonnet 4.6","description":"Claude Sonnet 4.6 is the 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 memory, polished document creation, and confident computer use for web QA and workflow automation.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["file-input","reasoning","tool-use","vision","explicit-caching","web-search"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024}],"temperature":true,"pricing":{"input_cache_read":"0.000000315","input_cache_write":"0.0000039375","input":"0.00000315","output":"0.00001575","web_search":"10.5"}},{"id":"anthropic/claude-sonnet-5","object":"model","created":1755815280,"released":1782691200,"owned_by":"anthropic","name":"Claude Sonnet 5","description":"Sonnet 5 is an upgrade to Sonnet 4.6, with gains across agentic coding and professional work. It builds on the strengths of previous Sonnet models, bringing top-tier intelligence at Sonnet pricing for coding, agents, and everyday professional work at scale.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","explicit-caching","file-input","vision","web-search"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input_cache_write":"0.000002625","input":"0.0000021","output":"0.0000105","web_search":"10.5"}},{"id":"anthropic/claude-opus-5","object":"model","created":1755815280,"released":1784851200,"owned_by":"anthropic","name":"Claude Opus 5","description":"Claude Opus 5 is the latest model in Anthropic's Opus family and a step-change improvement over Opus 4.8. It delivers major gains over Opus 4.8 in agentic coding, professional knowledge work, and long-horizon reasoning, and it is stronger per token across effort levels.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["tool-use","reasoning","vision","file-input","explicit-caching","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input_cache_write":"0.0000065625","input":"0.00000525","output":"0.00002625","web_search":"10.5"}},{"id":"deepseek/deepseek-v3.2","object":"model","created":1755815280,"released":1764547200,"owned_by":"deepseek","name":"DeepSeek V3.2","description":"DeepSeek-V3.2: Official successor to V3.2-Exp.","context_window":128000,"max_tokens":8000,"type":"language","tags":["tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000000294","input":"0.000000294","output":"0.000000441"}},{"id":"deepseek/deepseek-v4-pro-0813","object":"model","created":1755815280,"released":1786492800,"owned_by":"deepseek","name":"DeepSeek V4 Pro 0813","description":"This is the 8/13 updated weights version of DeepSeek V4 Pro.","context_window":1000000,"max_tokens":384000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000001386","input":"0.000001386","output":"0.000004158"}},{"id":"google/gemini-2.5-flash","object":"model","created":1755815280,"released":1742428800,"owned_by":"google","name":"Gemini 2.5 Flash","description":"Gemini 2.5 Flash is a thinking model that offers great, well-rounded capabilities. It is designed to offer a balance between price and performance with multimodal support and a 1M token context window.","context_window":1000000,"max_tokens":65536,"type":"language","tags":["file-input","reasoning","tool-use","vision","web-search","implicit-caching"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0,"max":24576}],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.000000315","output":"0.000002625","web_search":"36.75"}},{"id":"google/gemini-2.5-flash-image","object":"model","created":1755815280,"released":1756166400,"owned_by":"google","name":"Nano Banana (Gemini 2.5 Flash Image)","description":"Nano Banana (Gemini 2.5 Flash Image) is Google's first fully hybrid reasoning model, letting developers turn thinking on or off and set thinking budgets to balance quality, cost, and latency. Upgraded for rapid creative workflows, it can generate interleaved text and images and supports conversational, multi‑turn image editing in natural language. It’s also locale‑aware, enabling culturally and linguistically appropriate image generation for audiences worldwide.","context_window":32768,"max_tokens":65536,"type":"language","tags":["image-generation","web-search","vision"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.000000315","output":"0.000002625"}},{"id":"alibaba/qwen-3-30b","object":"model","created":1755815280,"released":1745798400,"owned_by":"alibaba","name":"Qwen3-30B-A3B","description":"Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support","context_window":40960,"max_tokens":16384,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.000000126","output":"0.000000525"}},{"id":"alibaba/qwen3.5-plus","object":"model","created":1755815280,"released":1771200000,"owned_by":"alibaba","name":"Qwen 3.5 Plus","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 task evaluations, the 3.5 series consistently demonstrates performance on par with state-of-the-art leading models. Compared to the 3 series, these models show a leap forward in both pure-text and multimodal capabilities.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["vision","file-input","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":81920}],"temperature":true,"pricing":{"input_cache_read":"0.000000042","input_cache_write":"0.000000525","input":"0.00000042","output":"0.00000252"}},{"id":"interfaze/interfaze-beta","object":"model","created":1755815280,"released":1759795200,"owned_by":"interfaze","name":"Interfaze Beta","description":"Interfaze is an AI model built on a new architecture that merges specialized DNN/CNN models with LLMs for developer tasks that require deterministic output and high consistency like OCR, scraping, classification, STT and more.","context_window":1000000,"max_tokens":32000,"type":"language","tags":["file-input","reasoning","tool-use","vision"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input":"0.000001575","output":"0.000003675"}},{"id":"kwaipilot/kat-coder-air-v2.5","object":"model","created":1755815280,"released":1783641600,"owned_by":"kwaipilot","name":"Kat Coder Air V2.5","description":"Fast response version of Kat Coder V2.5 optimized for Agent and Claw use cases.","context_window":256000,"max_tokens":80000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.0000001575","output":"0.00000063"}},{"id":"kwaipilot/kat-coder-pro-v2","object":"model","created":1755815280,"released":1774569600,"owned_by":"kwaipilot","name":"Kat Coder Pro V2","description":"A high-performance edition designed for complex enterprise projects and SaaS integration.","context_window":256000,"max_tokens":256000,"type":"language","tags":["tool-use","reasoning","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000063","input":"0.000000315","output":"0.00000126"}},{"id":"openai/gpt-5.1-codex-max","object":"model","created":1755815280,"released":1763510400,"owned_by":"openai","name":"GPT 5.1 Codex Max","description":"GPT‑5.1-Codex-Max is purpose-built for agentic coding.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000013125","input":"0.0000013125","output":"0.0000105","web_search":"14.7"}},{"id":"minimax/minimax-m2","object":"model","created":1755815280,"released":1761523200,"owned_by":"minimax","name":"MiniMax M2","description":"MiniMax-M2 redefines efficiency for agents. It is a compact, fast, and cost-effective MoE model (230 billion total parameters with 10 billion active parameters) built for elite performance in coding and agentic tasks, all while maintaining powerful general intelligence.","context_window":205000,"max_tokens":205000,"type":"language","tags":["reasoning","tool-use"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input_cache_write":"0.00000039375","input":"0.000000315","output":"0.00000126"}},{"id":"minimax/minimax-m2.1","object":"model","created":1755815280,"released":1766448000,"owned_by":"minimax","name":"MiniMax M2.1","description":"MiniMax 2.1 is MiniMax's latest model, optimized specifically for robustness in coding, tool use, instruction following, and long-horizon planning.","context_window":204800,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input_cache_write":"0.00000039375","input":"0.000000315","output":"0.00000126"}},{"id":"minimax/minimax-m2.7","object":"model","created":1755815280,"released":1773792000,"owned_by":"minimax","name":"Minimax M2.7","description":"M2.7 delivers outstanding performance in real-world software engineering, including end-to-end full project delivery, log analysis and bug troubleshooting, code security, machine learning, and more.","context_window":204800,"max_tokens":131000,"type":"language","tags":["tool-use","reasoning","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000063","input_cache_write":"0.00000039375","input":"0.000000315","output":"0.00000126"}},{"id":"nvidia/nemotron-3-ultra-550b-a55b","object":"model","created":1755815280,"released":1780531200,"owned_by":"nvidia","name":"Nemotron 3 Ultra","description":"A 550B parameter (55B active) open reasoning model from NVIDIA, built for long-running agent workflows. It uses a hybrid Mamba-Transformer MoE architecture and supports a 1M token context window.","context_window":1000000,"max_tokens":65000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000126","input":"0.00000063","output":"0.00000252"}},{"id":"meta/llama-4-scout","object":"model","created":1755815280,"released":1743811200,"owned_by":"meta","name":"Llama 4 Scout 17B Instruct","description":"Llama 4 Scout is the best multimodal model in the world in its class and is more powerful than our Llama 3 models, while fitting in a single H100 GPU. Additionally, Llama 4 Scout supports an industry-leading context window of up to 10M tokens.","context_window":128000,"max_tokens":8192,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.0000001785","output":"0.000000693"}},{"id":"mistral/devstral-2","object":"model","created":1755815280,"released":1765238400,"owned_by":"mistral","name":"Devstral 2","description":"An enterprise-grade text model that excels at using tools to explore codebases, editing multiple files, and powering software engineering agents.","context_window":256000,"max_tokens":256000,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.00000042","output":"0.0000021"}},{"id":"mistral/devstral-small-2","object":"model","created":1755815280,"released":1765238400,"owned_by":"mistral","name":"Devstral Small 2","description":"Our open source model that excels at using tools to explore codebases, editing multiple files, and powering software engineering agents.","context_window":256000,"max_tokens":256000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000105","output":"0.000000315"}},{"id":"mistral/magistral-medium","object":"model","created":1755815280,"released":1749513600,"owned_by":"mistral","name":"Magistral Medium 2509","description":"Complex thinking, backed by deep understanding, with transparent reasoning you can follow and verify. The model excels in maintaining high-fidelity reasoning across numerous languages, even when switching between languages mid-task.","context_window":128000,"max_tokens":64000,"type":"language","tags":["reasoning","tool-use","vision"],"temperature":true,"pricing":{"input":"0.0000021","output":"0.00000525"}},{"id":"openai/gpt-4o-mini","object":"model","created":1755815280,"released":1721260800,"owned_by":"openai","name":"GPT-4o mini","description":"GPT-4o mini from OpenAI is their most advanced and cost-efficient small model. It is multi-modal (accepting text or image inputs and outputting text) and has higher intelligence than gpt-3.5-turbo but is just as fast.","context_window":128000,"max_tokens":16384,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.00000007875","input":"0.0000001575","output":"0.00000063","web_search":"10.5"}},{"id":"openai/gpt-5.2-codex","object":"model","created":1755815280,"released":1766016000,"owned_by":"openai","name":"GPT 5.2 Codex","description":"GPT‑5.2-Codex is a version of GPT‑5.2⁠ further optimized for agentic coding in Codex, including improvements on long-horizon work through context compaction, stronger performance on large code changes like refactors and migrations, improved performance in Windows environments, and significantly stronger cybersecurity capabilities.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000018375","input":"0.0000018375","output":"0.0000147","web_search":"14.7"}},{"id":"openai/gpt-5.3-codex-fast","object":"model","created":1755815280,"released":1770249600,"owned_by":"openai","name":"GPT 5.3 Codex (Fast)","description":"GPT-5.3-Codex advances both the frontier coding performance of GPT‑5.2-Codex and the reasoning and professional knowledge capabilities of GPT‑5.2, together in one model, which is also 25% faster. This enables it to take on long-running tasks that involve research, tool use, and complex execution.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000003675","input":"0.000003675","output":"0.0000294","web_search":"10.5"}},{"id":"mistral/ministral-8b","object":"model","created":1755815280,"released":1729036800,"owned_by":"mistral","name":"Ministral 8B","description":"A more powerful model with faster, memory-efficient inference, ideal for complex workflows and demanding edge applications.","context_window":128000,"max_tokens":4000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.0000001575"}},{"id":"mistral/mistral-large-3","object":"model","created":1755815280,"released":1764633600,"owned_by":"mistral","name":"Mistral Large 3","description":"Mistral Large 3 2512 is Mistral’s most capable model to date. It has a sparse mixture-of-experts architecture with 41B active parameters (675B total).","context_window":256000,"max_tokens":256000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000525","output":"0.000001575"}},{"id":"mistral/mistral-medium","object":"model","created":1755815280,"released":1746576000,"owned_by":"mistral","name":"Mistral Medium 3.1","description":"Mistral Medium 3 delivers frontier performance while being an order of magnitude less expensive. For instance, the model performs at or above 90% of Claude Sonnet 3.7 on benchmarks across the board at a significantly lower cost.","context_window":128000,"max_tokens":64000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.00000042","output":"0.0000021"}},{"id":"mistral/mistral-medium-3.5","object":"model","created":1755815280,"released":1777420800,"owned_by":"mistral","name":"Mistral Medium Latest","description":"Mistral's frontier-class multimodal model optimized for agentic and coding use cases.","context_window":256000,"max_tokens":256000,"type":"language","tags":["reasoning","tool-use","vision"],"reasoning_options":[{"type":"effort","values":["none","high"]}],"temperature":true,"pricing":{"input":"0.000001575","output":"0.000007875"}},{"id":"mistral/mistral-nemo","object":"model","created":1755815280,"released":1721260800,"owned_by":"mistral","name":"Mistral Nemo 12B","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, Korean, Arabic, and Hindi. It supports function calling and is released under the Apache 2.0 license.","context_window":128000,"max_tokens":128000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.0000001575"}},{"id":"mistral/mistral-small","object":"model","created":1755815280,"released":1726531200,"owned_by":"mistral","name":"Mistral Small","description":"Mistral Small is the ideal choice for simple tasks that one can do in bulk - like Classification, Customer Support, or Text Generation. It offers excellent performance at an affordable price point.","context_window":32000,"max_tokens":4000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000105","output":"0.000000315"}},{"id":"mistral/pixtral-12b","object":"model","created":1755815280,"released":1726531200,"owned_by":"mistral","name":"Pixtral 12B 2409","description":"A 12B model with image understanding capabilities in addition to text.","context_window":128000,"max_tokens":4000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.0000001575"}},{"id":"moonshotai/kimi-k2.5","object":"model","created":1755815280,"released":1769385600,"owned_by":"moonshotai","name":"Kimi K2.5","description":"kimi-k2.5 is Kimi's most versatile model to date, featuring a native multimodal architecture that supports both visual and text input, thinking and non-thinking modes, and dialogue and agent tasks.","context_window":262114,"max_tokens":262114,"type":"language","tags":["reasoning","vision","tool-use","implicit-caching","video-input"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000105","input":"0.00000063","output":"0.00000315"}},{"id":"moonshotai/kimi-k3-fast","object":"model","created":1755815280,"released":1785110400,"owned_by":"moonshotai","name":"Kimi K3 Fast","description":"Fast version of Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window.","context_window":1000000,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision"],"temperature":false,"pricing":{"input_cache_read":"0.0000004725","input":"0.000004725","output":"0.000023625"}},{"id":"openai/gpt-5.4-mini-fast","object":"model","created":1755815280,"released":1773705600,"owned_by":"openai","name":"GPT 5.4 Mini (Fast)","description":"GPT-5.4 Mini brings the strengths of GPT-5.4 to a faster, more efficient model designed for high-volume workloads.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000001575","input":"0.000001575","output":"0.00000945","web_search":"10.5"}},{"id":"openai/gpt-5.4-nano","object":"model","created":1755815280,"released":1773705600,"owned_by":"openai","name":"GPT 5.4 Nano","description":"GPT-5.4 Nano is designed for tasks where speed and cost matter most like classification, data extraction, ranking, and sub-agents.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000021","input":"0.00000021","output":"0.0000013125","web_search":"10.5"}},{"id":"openai/gpt-oss-safeguard-20b","object":"model","created":1755815280,"released":1761696000,"owned_by":"openai","name":"GPT OSS Safeguard 20B","description":"OpenAI's first open weight reasoning model specifically trained for safety classification tasks. Fine-tuned from GPT-OSS, this model helps classify text content based on customizable policies, enabling bring-your-own-policy Trust & Safety AI where your own taxonomy, definitions, and thresholds guide classification decisions.","context_window":131072,"max_tokens":65536,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000003885","input":"0.00000007875","output":"0.000000315"}},{"id":"morph/morph-v3-fast","object":"model","created":1755815280,"released":1751846400,"owned_by":"morph","name":"Morph V3 Fast","description":"Morph offers a specialized AI model that applies code changes suggested by frontier models (like Claude or GPT-4o) to your existing code files FAST - 4500+ tokens/second. It acts as the final step in the AI coding workflow. Supports 16k input tokens and 16k output tokens.","context_window":81920,"max_tokens":16384,"type":"language","tags":[],"temperature":false,"pricing":{"input":"0.00000084","output":"0.00000126"}},{"id":"morph/morph-v3-large","object":"model","created":1755815280,"released":1751846400,"owned_by":"morph","name":"Morph V3 Large","description":"Morph offers a specialized AI model that applies code changes suggested by frontier models (like Claude or GPT-4o) to your existing code files FAST - 2500+ tokens/second. It acts as the final step in the AI coding workflow. Supports 16k input tokens and 16k output tokens.","context_window":81920,"max_tokens":16384,"type":"language","tags":[],"temperature":false,"pricing":{"input":"0.000000945","output":"0.000001995"}},{"id":"nvidia/nemotron-3-nano-30b-a3b","object":"model","created":1755815280,"released":1765756800,"owned_by":"nvidia","name":"Nemotron 3 Nano 30B A3B","description":"NVIDIA Nemotron 3 Nano is an open reasoning model optimized for fast, cost-efficient inference. Built with a hybrid MoE and Mamba architecture and trained on NVIDIA-curated synthetic reasoning data, it delivers strong multi-step reasoning with stable latency and predictable performance for agentic and production workloads.","context_window":262144,"max_tokens":262144,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.0000000525","output":"0.000000252"}},{"id":"openai/gpt-3.5-turbo","object":"model","created":1755815280,"released":1677628800,"owned_by":"openai","name":"GPT-3.5 Turbo","description":"OpenAI's most capable and cost effective model in the GPT-3.5 family optimized for chat purposes, but also works well for traditional completions tasks.","context_window":16385,"max_tokens":4096,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.000000525","output":"0.000001575"}},{"id":"openai/gpt-4-turbo","object":"model","created":1755815280,"released":1712620800,"owned_by":"openai","name":"GPT-4 Turbo","description":"gpt-4-turbo from OpenAI has broad general knowledge and domain expertise allowing it to follow complex instructions in natural language and solve difficult problems accurately. It has a knowledge cutoff of April 2023 and a 128,000 token context window.","context_window":128000,"max_tokens":4096,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.0000105","output":"0.0000315"}},{"id":"openai/gpt-4.1","object":"model","created":1755815280,"released":1744588800,"owned_by":"openai","name":"GPT-4.1","description":"GPT 4.1 is OpenAI's flagship model for complex tasks. It is well suited for problem solving across domains.","context_window":1047576,"max_tokens":32768,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input":"0.0000021","output":"0.0000084","web_search":"10.5"}},{"id":"openai/gpt-5-codex","object":"model","created":1755815280,"released":1757894400,"owned_by":"openai","name":"GPT-5-Codex","description":"GPT-5-Codex is a version of GPT-5 optimized for agentic coding tasks in Codex or similar environments.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.0000001365","input":"0.0000013125","output":"0.0000105","web_search":"14.7"}},{"id":"openai/gpt-5-mini","object":"model","created":1755815280,"released":1754524800,"owned_by":"openai","name":"GPT-5 mini","description":"GPT-5 mini is a cost optimized model that excels at reasoning/chat tasks. It offers an optimal balance between speed, cost, and capability.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.00000002625","input":"0.0000002625","output":"0.0000021","web_search":"10.5"}},{"id":"openai/gpt-5-nano","object":"model","created":1755815280,"released":1754524800,"owned_by":"openai","name":"GPT-5 nano","description":"GPT-5 nano is a high throughput model that excels at simple instruction or classification tasks.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.00000000525","input":"0.0000000525","output":"0.00000042","web_search":"10.5"}},{"id":"openai/gpt-5-pro","object":"model","created":1755815280,"released":1759708800,"owned_by":"openai","name":"GPT-5 pro","description":"GPT-5 pro uses more compute to think harder and provide consistently better answers. Since GPT-5 pro is designed to tackle tough problems, some requests may take several minutes to finish.","context_window":400000,"max_tokens":272000,"type":"language","tags":["file-input","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["high"]}],"temperature":true,"pricing":{"input":"0.00001575","output":"0.000126","web_search":"10.5"}},{"id":"alibaba/qwen-3.6-max-preview","object":"model","created":1755815280,"released":1776643200,"owned_by":"alibaba","name":"Qwen 3.6 Max Preview","description":"Compared with the previously released Qwen3-Max and Qwen3.6-Plus, this model features enhanced vibe coding abilities, more efficient coding agent execution, and significantly improved front-end development skills. Additionally, its long-tail knowledge retention has been further upgraded.","context_window":240000,"max_tokens":64000,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":131072}],"temperature":true,"pricing":{"input_cache_read":"0.000000273","input_cache_write":"0.00000170625","input":"0.000001365","output":"0.00000819"}},{"id":"alibaba/qwen3.6-plus","object":"model","created":1755815280,"released":1775088000,"owned_by":"alibaba","name":"Qwen 3.6 Plus","description":"The Qwen3.6 native vision-language Plus series models demonstrate exceptional performance on par with the current state-of-the-art models, with a significant improvement in overall results compared to the 3.5 series. The models have been markedly enhanced in code-related capabilities such as agentic coding, front-end programming, and Vibe coding, as well as in multi-modal general object recognition, OCR, and object localization.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["reasoning","tool-use","vision","file-input"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1,"max":131072}],"temperature":true,"pricing":{"input_cache_read":"0.000000105","input_cache_write":"0.00000065625","input":"0.000000525","output":"0.00000315"}},{"id":"openai/gpt-5.6-sol-fast","object":"model","created":1755815280,"released":1783555200,"owned_by":"openai","name":"GPT 5.6 Sol (Fast)","description":"GPT-5.6 Sol is the flagship of OpenAI's GPT-5.6 series, its most capable model for long-horizon agentic work across coding, biology, and cybersecurity.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000042","input_cache_write":"0.000002625","input":"0.0000042","output":"0.000021","web_search":"10.5"}},{"id":"openai/gpt-5.6-terra-fast","object":"model","created":1755815280,"released":1783555200,"owned_by":"openai","name":"GPT 5.6 Terra (Fast)","description":"GPT-5.6 Terra is a balanced GPT-5.6 model for everyday work, with performance comparable to the previous generation at half the cost.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["reasoning","web-search","file-input","tool-use","vision","implicit-caching","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000042","input_cache_write":"0.000002625","input":"0.0000042","output":"0.0000252","web_search":"10.5"}},{"id":"alibaba/qwen3-vl-235b-a22b-instruct","object":"model","created":1755815280,"released":1758585600,"owned_by":"alibaba","name":"Qwen3 VL 235B A22B Instruct","description":"The Qwen3 series VL models has been comprehensively upgraded in areas such as visual coding and spatial perception. Its visual perception and recognition capabilities have significantly improved, supporting the understanding of ultra-long videos, and its OCR functionality has undergone a major enhancement.","context_window":131072,"max_tokens":129024,"type":"language","tags":["tool-use","vision","implicit-caching"],"temperature":true,"pricing":{"input":"0.00000042","output":"0.00000168"}},{"id":"alibaba/qwen3-vl-instruct","object":"model","created":1755815280,"released":1758585600,"owned_by":"alibaba","name":"Qwen3 VL 235B A22B Instruct","description":"The Qwen3 series VL models has been comprehensively upgraded in areas such as visual coding and spatial perception. Its visual perception and recognition capabilities have significantly improved, supporting the understanding of ultra-long videos, and its OCR functionality has undergone a major enhancement.","context_window":131072,"max_tokens":129024,"type":"language","tags":["tool-use","vision","implicit-caching"],"temperature":true,"pricing":{"input":"0.00000042","output":"0.00000168"}},{"id":"anthropic/claude-opus-4.5","object":"model","created":1755815280,"released":1763942400,"owned_by":"anthropic","name":"Claude Opus 4.5","description":"Claude Opus 4.5 is Anthropic’s latest model in the Opus series, meant for demanding reasoning tasks and complex problem solving. This model has improvements in general intelligence and vision compared to previous iterations. In addition, it is suited for difficult coding tasks and agentic workflows, especially those with computer use and tool use, and can effectively handle context usage and external memory files.","context_window":200000,"max_tokens":64000,"type":"language","tags":["explicit-caching","file-input","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":1024}],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input_cache_write":"0.0000065625","input":"0.00000525","output":"0.00002625","web_search":"10.5"}},{"id":"anthropic/claude-opus-4.7","object":"model","created":1755815280,"released":1776297600,"owned_by":"anthropic","name":"Claude Opus 4.7","description":"Opus 4.7 builds on the coding and agentic strengths of Opus 4.6 with stronger performance on complex, multi-step tasks and more reliable agentic execution. It also brings improved performance on knowledge work, from drafting documents to building presentations and analyzing data.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["tool-use","reasoning","vision","file-input","explicit-caching","web-search"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input_cache_write":"0.0000065625","input":"0.00000525","output":"0.00002625","web_search":"10.5"}},{"id":"amazon/nova-2-lite","object":"model","created":1755815280,"released":1764633600,"owned_by":"amazon","name":"Nova 2 Lite","description":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text.","context_window":1000000,"max_tokens":1000000,"type":"language","tags":["file-input","reasoning","tool-use","vision"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000007875","input":"0.000000315","output":"0.000002625"}},{"id":"amazon/nova-lite","object":"model","created":1755815280,"released":1733184000,"owned_by":"amazon","name":"Nova Lite","description":"A very low cost multimodal model that is lightning fast for processing image, video, and text inputs.","context_window":300000,"max_tokens":8192,"type":"language","tags":["file-input","tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000063","output":"0.000000252"}},{"id":"amazon/nova-micro","object":"model","created":1755815280,"released":1733184000,"owned_by":"amazon","name":"Nova Micro","description":"A text-only model that delivers the lowest latency responses at very low cost.","context_window":128000,"max_tokens":8192,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.00000003675","output":"0.000000147"}},{"id":"anthropic/claude-opus-5-fast","object":"model","created":1755815280,"released":1784851200,"owned_by":"anthropic","name":"Claude Opus 5 (Fast)","description":"Claude Opus 5 is the latest model in Anthropic's Opus family and a step-change improvement over Opus 4.8. It delivers major gains over Opus 4.8 in agentic coding, professional knowledge work, and long-horizon reasoning, and it is stronger per token across effort levels.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["tool-use","reasoning","vision","file-input","explicit-caching","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.00000105","input_cache_write":"0.000013125","input":"0.0000105","output":"0.0000525","web_search":"10.5"}},{"id":"anthropic/claude-sonnet-4","object":"model","created":1755815280,"released":1747872000,"owned_by":"anthropic","name":"Claude Sonnet 4","description":"Claude Sonnet 4 balances impressive performance for coding with the right speed and cost for high-volume use cases: Coding: Handle everyday development tasks with enhanced performance-power code reviews, bug fixes, API integrations, and feature development with immediate feedback loops.","context_window":1000000,"max_tokens":8192,"type":"language","tags":["file-input","reasoning","tool-use","vision","explicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000315","input_cache_write":"0.0000039375","input":"0.00000315","output":"0.00001575"}},{"id":"anthropic/claude-opus-4.8","object":"model","created":1755815280,"released":1779926400,"owned_by":"anthropic","name":"Claude Opus 4.8","description":"Opus 4.8 is a focused upgrade to Opus 4.7 and is Anthropic's best generally available model for coding, agentic tasks, and enterprise workflows. It builds on the strengths of previous Opus models with stronger performance on complex, multi-step coding tasks. Anthropic recommends using it on long-horizon coding and agentic tasks. It is also stronger on professional work, including document drafting, data analysis, and presentations.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["tool-use","reasoning","vision","file-input","explicit-caching","web-search","fast"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input_cache_write":"0.0000065625","input":"0.00000525","output":"0.00002625","web_search":"10.5"}},{"id":"arcee-ai/trinity-mini","object":"model","created":1755815280,"released":1764547200,"owned_by":"arcee-ai","name":"Trinity Mini","description":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model, engineered for efficient inference over long contexts with robust function calling and multi-step agent workflows.","context_window":131072,"max_tokens":131072,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.00000004725","output":"0.0000001575"}},{"id":"bytedance/seed-1.8","object":"model","created":1755815280,"released":1756684800,"owned_by":"bytedance","name":"Bytedance Seed 1.8","description":"Bytedance Seed 1.8 features stronger multimodal understanding and agent capabilities. The model delivers superior performance across a wide range of complex real-world tasks, helping enterprises create greater value.","context_window":256000,"max_tokens":64000,"type":"language","tags":["reasoning","tool-use","vision"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000000525","input":"0.0000002625","output":"0.0000021"}},{"id":"openai/o3","object":"model","created":1755815280,"released":1744761600,"owned_by":"openai","name":"o3","description":"OpenAI's o3 is their most powerful reasoning model, setting new state-of-the-art benchmarks in coding, math, science, and visual perception. It excels at complex queries requiring multi-faceted analysis, with particular strength in analyzing images, charts, and graphics.","context_window":200000,"max_tokens":100000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.000000525","input":"0.0000021","output":"0.0000084","web_search":"10.5"}},{"id":"cohere/command-a","object":"model","created":1755815280,"released":1741824000,"owned_by":"cohere","name":"Command A","description":"Command A is Cohere's most performant model to date, excelling at tool use, agents, retrieval augmented generation (RAG), and multilingual use cases. Command A has a context length of 256K, only requires two GPUs to run, and has 150% higher throughput compared to Command R+ 08-2024.","context_window":256000,"max_tokens":8000,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.000002625","output":"0.0000105"}},{"id":"deepseek/deepseek-r1","object":"model","created":1755815280,"released":1737331200,"owned_by":"deepseek","name":"DeepSeek-R1","description":"DeepSeek-R1 provides customers a state-of-the-art reasoning model, optimized for general reasoning tasks, math, science, and code generation.","context_window":128000,"max_tokens":8192,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input":"0.0000014175","output":"0.00000567"}},{"id":"deepseek/deepseek-v3","object":"model","created":1755815280,"released":1735171200,"owned_by":"deepseek","name":"DeepSeek V3 0324","description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","context_window":163840,"max_tokens":163840,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input_cache_read":"0.00000014175","input":"0.0000002835","output":"0.000001176"}},{"id":"deepseek/deepseek-v3.1","object":"model","created":1755815280,"released":1755734400,"owned_by":"deepseek","name":"DeepSeek V3.1","description":"DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoint through a two-phase long context extension approach, following the methodology outlined in the original DeepSeek-V3 report. We have expanded our dataset by collecting additional long documents and substantially extending both training phases. The 32K extension phase has been increased 10-fold to 630B tokens, while the 128K extension phase has been extended by 3.3x to 209B tokens. Additionally, DeepSeek-V3.1 is trained using the UE8M0 FP8 scale data format to ensure compatibility with microscaling data formats.","context_window":163840,"max_tokens":128000,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.0000001365","input":"0.0000002625","output":"0.0000009975"}},{"id":"deepseek/deepseek-v3.1-terminus","object":"model","created":1755815280,"released":1758499200,"owned_by":"deepseek","name":"DeepSeek V3.1 Terminus","description":"DeepSeek-V3.1-Terminus delivers more stable & reliable outputs across benchmarks compared to the previous version and addresses user feedback (i.e. language consistency and agent upgrades).","context_window":131072,"max_tokens":65536,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.00000014175","input":"0.0000002835","output":"0.00000105"}},{"id":"openai/o3-fast","object":"model","created":1755815280,"released":1744761600,"owned_by":"openai","name":"o3 (Fast)","description":"OpenAI's o3 is their most powerful reasoning model, setting new state-of-the-art benchmarks in coding, math, science, and visual perception. It excels at complex queries requiring multi-faceted analysis, with particular strength in analyzing images, charts, and graphics.","context_window":200000,"max_tokens":100000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.00000091875","input":"0.000003675","output":"0.0000147","web_search":"10.5"}},{"id":"google/gemini-2.5-pro","object":"model","created":1755815280,"released":1742428800,"owned_by":"google","name":"Gemini 2.5 Pro","description":"Gemini 2.5 Pro is our most advanced reasoning Gemini model, capable of solving complex problems. Gemini 2.5 Pro can comprehend vast datasets and challenging problems from different information sources, including text, audio, images, video, and even entire code repositories.","context_window":1048576,"max_tokens":65536,"type":"language","tags":["file-input","reasoning","tool-use","vision","web-search","implicit-caching"],"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"temperature":true,"pricing":{"input_cache_read":"0.00000013125","input":"0.0000013125","output":"0.0000105","web_search":"36.75"}},{"id":"google/gemini-3-pro-image","object":"model","created":1755815280,"released":1756684800,"owned_by":"google","name":"Nano Banana Pro (Gemini 3 Pro Image)","description":"Nano Banana Pro (Gemini 3 Pro Image) builds on Nano Banana's generation capabilities into a new era of studio-quality, functional design to help you create and edit high-fidelity, production-ready visuals with unparalleled precision and control. Improvements include enhanced world knowledge and reasoning, dynamic text and translation, and studio level controls.","context_window":65536,"max_tokens":32768,"type":"language","tags":["image-generation","web-search","implicit-caching","vision"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000021","output":"0.0000126","web_search":"14.7"}},{"id":"google/gemini-3.1-flash-image-preview","object":"model","created":1755815280,"released":1772064000,"owned_by":"google","name":"Gemini 3.1 Flash Image Preview (Nano Banana 2)","description":"Gemini 3.1 Flash Image (Nano Banana 2) is optimized for image understanding and generation and offers a balance of price and performance.","context_window":131072,"max_tokens":32768,"type":"language","tags":["image-generation","implicit-caching","reasoning","vision","web-search"],"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000000525","input":"0.000000525","output":"0.00000315","web_search":"14.7"}},{"id":"google/gemini-3.1-flash-lite-image","object":"model","created":1755815280,"released":1782777600,"owned_by":"google","name":"Gemini 3.1 Flash Lite Image (Nano Banana 2 Lite)","description":"Gemini 3.1 Flash-Lite Image (Nano Banana 2 Lite) is Google's fastest image generation model enabling rapid creation and iteration.","context_window":65536,"max_tokens":4096,"type":"language","tags":["image-generation","implicit-caching","reasoning","vision","web-search"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.0000002625","output":"0.000001575","web_search":"14.7"}},{"id":"google/gemini-3.5-flash","object":"model","created":1755815280,"released":1779148800,"owned_by":"google","name":"Gemini 3.5 Flash","description":"Google's latest model, highly optimized for coding proficiency and parallel agentic execution loops. Defaults to medium thinking effort for faster and more cost-efficient responses.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["reasoning","file-input","vision","tool-use","web-search","implicit-caching","video-input"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000001575","input":"0.000001575","output":"0.00000945","web_search":"14.7"}},{"id":"google/gemini-3.6-flash","object":"model","created":1755815280,"released":1784592000,"owned_by":"google","name":"Gemini 3.6 Flash","description":"Gemini 3.6 Flash delivers higher quality across coding, agentic workflows, and web development with reduced token consumption and fewer model calls compared to previous model iterations.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["reasoning","file-input","tool-use","vision","implicit-caching","web-search","video-input"],"temperature":true,"pricing":{"input_cache_read":"0.00000007875","input":"0.0000007875","output":"0.0000039375","web_search":"14.7"}},{"id":"google/gemini-omni-flash-preview","object":"model","created":1755815280,"released":1782777600,"owned_by":"google","name":"Gemini Omni Flash Preview","description":"Gemini Omni Flash (Preview) is a multimodal model designed for video, image, and text tasks. It is optimized for video generation, offering video output alongside text responses in a single model.","context_window":1000000,"max_tokens":57920,"type":"language","tags":["file-input","reasoning","vision","video-generation"],"temperature":true,"pricing":{"audio_input_token_cost":"0.000001575","input":"0.000001575","output":"0.00000945"}},{"id":"google/gemma-4-31b-it","object":"model","created":1755815280,"released":1775088000,"owned_by":"google","name":"Gemma 4 31B IT","description":"Gemma 4 31B is engineered to tackle the most demanding enterprise workloads and complex reasoning tasks. With an expansive 256K-token context window, the 31B model can effortlessly ingest entire codebases, and massive sets of images in a single prompt.","context_window":262144,"max_tokens":131072,"type":"language","tags":["file-input","reasoning","tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000147","output":"0.00000042"}},{"id":"inception/mercury-2","object":"model","created":1755815280,"released":1771891200,"owned_by":"inception","name":"Mercury 2","description":"A diffusion-based reasoning LLM that generates text via parallel refinement (not token-by-token), delivering real-time latency with ~1k tokens/sec plus 128K context and built-in tool/JSON support.","context_window":128000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000002625","input":"0.0000002625","output":"0.0000007875"}},{"id":"inception/mercury-coder-small","object":"model","created":1755815280,"released":1740528000,"owned_by":"inception","name":"Mercury Coder Small Beta","description":"Mercury Coder Small is ideal for code generation, debugging, and refactoring tasks with minimal latency.","context_window":32000,"max_tokens":16384,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.0000002625","output":"0.00000105"}},{"id":"inclusionai/ling-3.0-flash","object":"model","created":1755815280,"released":1785974400,"owned_by":"inclusionai","name":"Ling 3.0 Flash","description":"Ling 3.0 Flash is designed with token efficiency and production-scale agentic inference as key priorities, enabling developers to complete more useful work within constrained token, latency, and serving-cost budgets.","context_window":256000,"max_tokens":32000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000000126","input":"0.000000063","output":"0.000000189"}},{"id":"kwaipilot/kat-coder-pro-v2.5","object":"model","created":1755815280,"released":1783641600,"owned_by":"kwaipilot","name":"Kat Coder Pro V2.5","description":"KAT-Coder-V2.5 is a coding-focused agentic model trained to act autonomously inside real, executable repositories rather than as a single-turn code generator.","context_window":256000,"max_tokens":80000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"temperature":true,"pricing":{"input_cache_read":"0.0000001575","input":"0.000000777","output":"0.000003108"}},{"id":"meta/llama-3.1-70b","object":"model","created":1755815280,"released":1721692800,"owned_by":"meta","name":"Llama 3.1 70B Instruct","description":"An update to Meta Llama 3 70B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities.","context_window":128000,"max_tokens":8192,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.000000756","output":"0.000000756"}},{"id":"meta/llama-3.1-8b","object":"model","created":1755815280,"released":1721692800,"owned_by":"meta","name":"Llama 3.1 8B Instruct","description":"An update to Meta Llama 3 8B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities.","context_window":128000,"max_tokens":8192,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.000000231","output":"0.000000231"}},{"id":"meta/llama-3.3-70b","object":"model","created":1755815280,"released":1733443200,"owned_by":"meta","name":"Llama 3.3 70B Instruct","description":"Where performance meets efficiency. This model supports high-performance conversational AI designed for content creation, enterprise applications, and research, offering advanced language understanding capabilities, including text summarization, classification, sentiment analysis, and code generation.","context_window":128000,"max_tokens":8192,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.000000756","output":"0.000000756"}},{"id":"tencent/hy3","object":"model","created":1755815280,"released":1783296000,"owned_by":"tencent","name":"Hy3","description":"Tencent Hy3 is an open-source Mixture-of-Experts (MoE) large language model developed by Tencent's Hunyuan team. It has 295B total parameters, 21B active parameters, a 3.8B-parameter MTP layer, and a 256K context window. Hy3 is designed for agentic workflows, coding, reasoning, long-context understanding, tool-use scenarios, document analysis, and structured generation.\n\nThe model improves on Hy3 Preview with stronger tool-call reliability, better output-format stability, improved long-context retention, and lower hallucination rates in Tencent's internal evaluations.","context_window":256000,"max_tokens":128000,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"temperature":true,"pricing":{"input_cache_read":"0.00000003465","input":"0.0000001386","output":"0.0000005544"}},{"id":"minimax/minimax-m2.1-lightning","object":"model","created":1755815280,"released":1766448000,"owned_by":"minimax","name":"MiniMax M2.1 Lightning","description":"MiniMax-M2.1-lightning is a faster version of MiniMax-M2.1, offering the same performance but with significantly higher throughput (output speed ~100 TPS, MiniMax-M2 output speed ~60 TPS).","context_window":204800,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input_cache_write":"0.00000039375","input":"0.000000315","output":"0.00000252"}},{"id":"minimax/minimax-m2.5","object":"model","created":1755815280,"released":1770854400,"owned_by":"minimax","name":"MiniMax M2.5","description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. It is capable of handling the entire development process of various complex systems. It covers full-stack projects across multiple platforms including Web, Android, iOS, Windows, and Mac, encompassing server-side APIs, functional logic, and databases.","context_window":204800,"max_tokens":131000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input_cache_write":"0.00000039375","input":"0.000000315","output":"0.00000126"}},{"id":"minimax/minimax-m2.5-highspeed","object":"model","created":1755815280,"released":1770854400,"owned_by":"minimax","name":"MiniMax M2.5 High Speed","description":"M2.5 highspeed: Same performance, faster and more agile (output speed approximately 100 tps)","context_window":204800,"max_tokens":131000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input_cache_write":"0.00000039375","input":"0.00000063","output":"0.00000252"}},{"id":"minimax/minimax-m2.7-highspeed","object":"model","created":1755815280,"released":1773792000,"owned_by":"minimax","name":"MiniMax M2.7 High Speed","description":"M2.7 Highspeed: Same performance, faster and more agile (output speed approximately 100 tps)","context_window":204800,"max_tokens":131100,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000063","input_cache_write":"0.00000039375","input":"0.00000063","output":"0.00000252"}},{"id":"sakana/namazu","object":"model","created":1755815280,"released":1785715200,"owned_by":"sakana","name":"Sakana Namazu","description":"Sakana Namazu is a Japanese-specialized LLM that combines a deep understanding of Japanese culture and business customs with high-performance language capabilities. Built on the open model Kimi K2.6 and refined with Sakana AI's in-house data for Japanese language and business workflows, it handles complex tasks using web search and code execution. Unlike Fugu, which orchestrates multiple frontier models, Sakana Namazu provides a single in-house model as an API.","context_window":256000,"max_tokens":256000,"type":"language","tags":["vision","file-input","implicit-caching","reasoning","structured-output","tool-use","web-search"],"temperature":true,"pricing":{"input_cache_read":"0.0000001575","input":"0.0000009975","output":"0.0000042"}},{"id":"google/gemini-2.5-flash-lite","object":"model","created":1755815280,"released":1750118400,"owned_by":"google","name":"Gemini 2.5 Flash Lite","description":"Gemini 2.5 Flash-Lite is a balanced, low-latency model with configurable thinking budgets and tool connectivity (e.g., Google Search grounding and code execution). It supports multimodal input and offers a 1M-token context window.","context_window":1048576,"max_tokens":65536,"type":"language","tags":["file-input","reasoning","tool-use","vision","web-search","implicit-caching"],"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":512,"max":24576}],"temperature":true,"pricing":{"input_cache_read":"0.0000000105","input":"0.000000105","output":"0.00000042","web_search":"36.75"}},{"id":"google/gemini-3.1-pro-preview","object":"model","created":1755815280,"released":1771459200,"owned_by":"google","name":"Gemini 3.1 Pro Preview","description":"This model improves upon Gemini 2.5 Pro and is catered towards challenging tasks, especially those involving complex reasoning or agentic workflows. Improvements highlighted include use cases for coding, multi-step function calling, planning, reasoning, deep knowledge tasks, and instruction following.","context_window":1000000,"max_tokens":64000,"type":"language","tags":["file-input","tool-use","reasoning","vision","web-search","implicit-caching"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000021","output":"0.0000126","web_search":"14.7"}},{"id":"google/gemma-4-26b-a4b-it","object":"model","created":1755815280,"released":1775088000,"owned_by":"google","name":"Google Gemma 4 26B A4B","description":"Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on small models) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages.","context_window":262144,"max_tokens":131072,"type":"language","tags":["file-input","reasoning","tool-use","vision","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.00000001575","input":"0.0000001575","output":"0.00000063"}},{"id":"spacexai/grok-4.3","object":"model","created":1755815280,"released":1777507200,"owned_by":"spacexai","name":"Grok 4.3","description":"Grok 4.3 is a new model matching the scale of Grok 4.20 with an improved architecture and a December 2025 knowledge cutoff.","context_window":1000000,"max_tokens":1000000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","web-search"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000013125","output":"0.000002625","web_search":"5.25"}},{"id":"mistral/magistral-small","object":"model","created":1755815280,"released":1749513600,"owned_by":"mistral","name":"Magistral Small 2509","description":"Complex thinking, backed by deep understanding, with transparent reasoning you can follow and verify. The model excels in maintaining high-fidelity reasoning across numerous languages, even when switching between languages mid-task.","context_window":128000,"max_tokens":64000,"type":"language","tags":["reasoning","tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000525","output":"0.000001575"}},{"id":"mistral/ministral-14b","object":"model","created":1755815280,"released":1764633600,"owned_by":"mistral","name":"Ministral 14B","description":"Ministral 3 14B is the largest model in the Ministral 3 family, offering state-of-the-art capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart. Optimized for local deployment, it delivers high performance across diverse hardware, including local setups.","context_window":256000,"max_tokens":256000,"type":"language","tags":["file-input","tool-use","vision"],"temperature":true,"pricing":{"input":"0.00000021","output":"0.00000021"}},{"id":"moonshotai/kimi-k2-thinking","object":"model","created":1755815280,"released":1762387200,"owned_by":"moonshotai","name":"Kimi K2 Thinking","description":"Kimi K2 Thinking is an advanced open-source thinking model by Moonshot AI. It can execute up to 200 – 300 sequential tool calls without human interference, reasoning coherently across hundreds of steps to solve complex problems. Built as a thinking agent, it reasons step by step while using tools, achieving state-of-the-art performance on Humanity's Last Exam (HLE), BrowseComp, and other benchmarks, with major gains in reasoning, agentic search, coding, writing, and general capabilities.","context_window":216144,"max_tokens":216144,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.00000014805","input":"0.0000004935","output":"0.0000021"}},{"id":"moonshotai/kimi-k2.6","object":"model","created":1755815280,"released":1776643200,"owned_by":"moonshotai","name":"Kimi K2.6","description":"Kimi K2.6 demonstrates particularly strong performance in long-horizon coding tasks and produces professional-grade design with code and vision.","context_window":262000,"max_tokens":262000,"type":"language","tags":["implicit-caching","reasoning","tool-use","vision","video-input"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000168","input":"0.0000009975","output":"0.0000042"}},{"id":"arcee-ai/trinity-large-thinking","object":"model","created":1755815280,"released":1775001600,"owned_by":"arcee-ai","name":"Trinity Large Thinking","description":"Trinity-Large-Thinking is a reasoning-optimized variant of Arcee AI's Trinity-Large family — a 398B-parameter sparse Mixture-of-Experts (MoE) model with approximately 13B active parameters per token. Built on Trinity-Large-Base and post-trained with extended chain-of-thought reasoning and agentic RL, Trinity-Large-Thinking delivers state-of-the-art performance on agentic benchmarks while maintaining strong general capabilities.","context_window":262100,"max_tokens":80000,"type":"language","tags":["reasoning","tool-use"],"temperature":true,"pricing":{"input":"0.0000002625","output":"0.000000945"}},{"id":"moonshotai/kimi-k2.7-code","object":"model","created":1755815280,"released":1781222400,"owned_by":"moonshotai","name":"Kimi K2.7 Code","description":"Kimi-K2.7-Code is a coding model from Moonshot AI. It has  improved coding & agent performance over K2.6, more reasoning efficiency with less overthinking, and improved instruction following for long-horizon coding.","context_window":256000,"max_tokens":32768,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","video-input"],"temperature":true,"pricing":{"input_cache_read":"0.0000001995","input":"0.0000009975","output":"0.0000042"}},{"id":"moonshotai/kimi-k2.7-code-highspeed","object":"model","created":1755815280,"released":1781481600,"owned_by":"moonshotai","name":"Kimi K2.7 Code High Speed","description":"Kimi K2.7 Code HighSpeed is the high-speed version of Kimi K2.7 Code, the same model as Kimi K2.7 Code, but with an output speed of approximately 180 Tokens/s and up to 260 Tokens/s in short context scenarios, delivering a more extreme coding experience.","context_window":262144,"max_tokens":32768,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","video-input"],"temperature":true,"pricing":{"input_cache_read":"0.000000399","input":"0.000001995","output":"0.0000084"}},{"id":"nvidia/nemotron-3-super-120b-a12b","object":"model","created":1755815280,"released":1773187200,"owned_by":"nvidia","name":"NVIDIA Nemotron 3 Super 120B A12B","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. It delivers up to 7x higher throughput, providing fast, cost-efficient inference for agentic tasks. Additionally, a long context window gives the model long-term memory, preventing AI agents from losing focus on long, multi-step tasks and ensuring high-accuracy results. Fully open with weights, datasets, and recipes, Super allows easy customization and secure deployment anywhere.","context_window":256000,"max_tokens":32000,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.0000006825"}},{"id":"nvidia/nemotron-nano-12b-v2-vl","object":"model","created":1755815280,"released":1761609600,"owned_by":"nvidia","name":"Nvidia Nemotron Nano 12B V2 VL","description":"The model is an auto-regressive vision language model that uses an optimized transformer architecture. The model enables multi-image reasoning and video understanding, along with strong document intelligence, visual Q&A and summarization capabilities.","context_window":131072,"max_tokens":131072,"type":"language","tags":["vision","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.00000021","output":"0.00000063"}},{"id":"mistral/ministral-3b","object":"model","created":1755815280,"released":1729036800,"owned_by":"mistral","name":"Ministral 3B","description":"A compact, efficient model for on-device tasks like smart assistants and local analytics, offering low-latency performance.","context_window":128000,"max_tokens":4000,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000105","output":"0.000000105"}},{"id":"openai/gpt-5.1-codex","object":"model","created":1755815280,"released":1762905600,"owned_by":"openai","name":"GPT-5.1-Codex","description":"GPT-5.1-Codex is a version of GPT-5.1 optimized for agentic coding tasks in Codex or similar environments.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000001365","input":"0.0000013125","output":"0.0000105","web_search":"14.7"}},{"id":"openai/gpt-5.1-codex-mini","object":"model","created":1755815280,"released":1762905600,"owned_by":"openai","name":"GPT 5.1 Codex Mini","description":"GPT-5.1 Codex mini is a smaller, faster, and cheaper version of GPT-5.1 Codex.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.0000002625","output":"0.0000021","web_search":"14.7"}},{"id":"openai/gpt-5.1-thinking","object":"model","created":1755815280,"released":1762905600,"owned_by":"openai","name":"GPT 5.1 Thinking","description":"An upgraded version of GPT-5 that adapts thinking time more precisely to the question to spend more time on complex questions and respond more quickly to simpler tasks.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000013125","input":"0.0000013125","output":"0.0000105","web_search":"10.5"}},{"id":"alibaba/qwen3.8-27b","object":"model","created":1755815280,"released":1786665600,"owned_by":"alibaba","name":"Qwen3.8 27B","description":"Built on the architectural foundation of Qwen3.5, Qwen3.8 delivers substantial gains across coding, professional work, research, and long-horizon agentic tasks. Qwen3.8-27B brings these advances to a compact, deployment-friendly dense model: a native vision-language model that understands images and videos, with flexible thinking control, designed to carry complex, multi-step tasks through to completion with greater reliability.","context_window":1000000,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use","vision","implicit-caching","video-input","file-input"],"temperature":true,"pricing":{"input_cache_read":"0.0000001155","input":"0.0000005775","output":"0.000003465"}},{"id":"alibaba/qwen3.8-max","object":"model","created":1755815280,"released":1785628800,"owned_by":"alibaba","name":"Qwen 3.8 Max","description":"Qwen 3.8 Max is a 2.4-trillion-parameter MoE model delivering a comprehensive leap in coding and professional work. Autonomously codes and delivers complete projects spanning 10+ days. Handles hundreds of specialized tasks across legal, financial, design, and other professional domains, producing production-grade results end-to-end in a single conversation. Native visual understanding runs through the full cycle of planning, execution, and verification, enabling deep semantic analysis of ultra-long documents and extended video content. In long-horizon tasks, plans autonomously, iterates through closed feedback loops, and continuously evolves.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"temperature":true,"pricing":{"input_cache_read":"0.0000002625","input_cache_write":"0.000002625","input":"0.0000021","output":"0.0000063"}},{"id":"nvidia/nemotron-nano-9b-v2","object":"model","created":1755815280,"released":1755475200,"owned_by":"nvidia","name":"Nvidia Nemotron Nano 9B V2","description":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be controlled via a system prompt. If the user prefers the model to provide its final answer without intermediate reasoning traces, it can be configured to do so.\\","context_window":131072,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.000000063","output":"0.0000002415"}},{"id":"alibaba/qwen3-max-thinking","object":"model","created":1755815280,"released":1769126400,"owned_by":"alibaba","name":"Qwen 3 Max Thinking","description":"Compared with the snapshot as of September 23, 2025, the Qwen-3 series Max model in this release achieves an effective integration of thinking and non-thinking modes, resulting in a comprehensive and substantial improvement in the model’s overall performance. In thinking mode, the model simultaneously supports web search, web information extraction, and a code interpreter tool, enabling it to tackle more complex and challenging problems with greater accuracy by leveraging external tools while engaging in slow, deliberative reasoning. This version is based on a snapshot taken on January 23, 2026.","context_window":256000,"max_tokens":65536,"type":"language","tags":["reasoning","tool-use"],"reasoning_options":[{"type":"budget_tokens","min":1,"max":81920}],"temperature":true,"pricing":{"input_cache_read":"0.000000252","input":"0.00000126","output":"0.0000063"}},{"id":"openai/gpt-5-fast","object":"model","created":1755815280,"released":1754524800,"owned_by":"openai","name":"GPT-5 (Fast)","description":"GPT-5 is OpenAI's flagship language model that excels at complex reasoning, broad real-world knowledge, code-intensive, and multi-step agentic tasks.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.0000002625","input":"0.000002625","output":"0.000021","web_search":"10.5"}},{"id":"openai/gpt-5-mini-fast","object":"model","created":1755815280,"released":1754524800,"owned_by":"openai","name":"GPT-5 mini (Fast)","description":"GPT-5 mini is a cost optimized model that excels at reasoning/chat tasks. It offers an optimal balance between speed, cost, and capability.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.00000004725","input":"0.0000004725","output":"0.00000378","web_search":"10.5"}},{"id":"openai/gpt-5.1-thinking-fast","object":"model","created":1755815280,"released":1762905600,"owned_by":"openai","name":"GPT 5.1 Thinking (Fast)","description":"An upgraded version of GPT-5 that adapts thinking time more precisely to the question to spend more time on complex questions and respond more quickly to simpler tasks.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000002625","input":"0.000002625","output":"0.000021","web_search":"10.5"}},{"id":"openai/gpt-5.2-fast","object":"model","created":1755815280,"released":1765411200,"owned_by":"openai","name":"GPT 5.2 (Fast)","description":"GPT-5.2 is OpenAI's best general-purpose model, part of the GPT-5 flagship model family. It's their most intelligent model yet for both general and agentic tasks.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000003675","input":"0.000003675","output":"0.0000294","web_search":"10.5"}},{"id":"openai/gpt-5.2-pro","object":"model","created":1755815280,"released":1765411200,"owned_by":"openai","name":"GPT 5.2 ","description":"Version of GPT-5.2 that produces smarter and more precise responses.","context_window":400000,"max_tokens":128000,"type":"language","tags":["tool-use","vision","reasoning","file-input","web-search"],"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"temperature":true,"pricing":{"input":"0.00002205","output":"0.0001764","web_search":"10.5"}},{"id":"openai/gpt-4.1-mini-fast","object":"model","created":1755815280,"released":1747180800,"owned_by":"openai","name":"GPT-4.1 mini (Fast)","description":"GPT 4.1 mini provides a balance between intelligence, speed, and cost that makes it an attractive model for many use cases.","context_window":1047576,"max_tokens":32768,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.00000018375","input":"0.000000735","output":"0.00000294","web_search":"10.5"}},{"id":"openai/gpt-5.4","object":"model","created":1755815280,"released":1772668800,"owned_by":"openai","name":"GPT 5.4","description":"GPT-5.4 is OpenAI's best general-purpose model, part of the GPT-5 flagship model family. It's their most intelligent model yet for both general and agentic tasks.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000002625","input":"0.000002625","output":"0.00001575","web_search":"10.5"}},{"id":"openai/gpt-5.4-fast","object":"model","created":1755815280,"released":1772668800,"owned_by":"openai","name":"GPT 5.4 (Fast)","description":"GPT-5.4 is OpenAI's best general-purpose model, part of the GPT-5 flagship model family. It's their most intelligent model yet for both general and agentic tasks.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input":"0.00000525","output":"0.0000315","web_search":"10.5"}},{"id":"openai/gpt-5.6-luna","object":"model","created":1755815280,"released":1783555200,"owned_by":"openai","name":"GPT 5.6 Luna","description":"GPT-5.6 Luna is a fast, affordable GPT-5.6 model that brings strong capability at the lowest cost in the series.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["reasoning","file-input","tool-use","vision","implicit-caching","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000021","input_cache_write":"0.0000002625","input":"0.00000021","output":"0.00000126","web_search":"10.5"}},{"id":"openai/gpt-oss-120b","object":"model","created":1755815280,"released":1754352000,"owned_by":"openai","name":"GPT OSS 120B","description":"Extremely capable general-purpose LLM with strong, controllable reasoning capabilities","context_window":131072,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input":"0.000000105","output":"0.000000525"}},{"id":"openai/gpt-oss-20b","object":"model","created":1755815280,"released":1754352000,"owned_by":"openai","name":"GPT OSS 20B","description":"A compact, open-weight language model optimized for low-latency and resource-constrained environments, including local and edge deployments.","context_window":131072,"max_tokens":8192,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input":"0.0000000525","output":"0.00000021"}},{"id":"openai/o3-pro","object":"model","created":1755815280,"released":1749513600,"owned_by":"openai","name":"o3 Pro","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 better answers.","context_window":200000,"max_tokens":100000,"type":"language","tags":["reasoning","vision","file-input","tool-use","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input":"0.000021","output":"0.000084","web_search":"10.5"}},{"id":"moonshotai/kimi-k3","object":"model","created":1755815280,"released":1784160000,"owned_by":"moonshotai","name":"Kimi K3","description":"Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window.","context_window":1000000,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","video-input"],"temperature":false,"pricing":{"input_cache_read":"0.000000315","input":"0.00000315","output":"0.00001575"}},{"id":"openai/gpt-4.1-mini","object":"model","created":1755815280,"released":1747180800,"owned_by":"openai","name":"GPT-4.1 mini","description":"GPT 4.1 mini provides a balance between intelligence, speed, and cost that makes it an attractive model for many use cases.","context_window":1047576,"max_tokens":32768,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.000000105","input":"0.00000042","output":"0.00000168","web_search":"10.5"}},{"id":"openai/gpt-4.1-nano-fast","object":"model","created":1755815280,"released":1744588800,"owned_by":"openai","name":"GPT-4.1 nano (Fast)","description":"GPT-4.1 nano is the fastest, most cost-effective GPT 4.1 model.","context_window":1047576,"max_tokens":32768,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.0000000525","input":"0.00000021","output":"0.00000084","web_search":"10.5"}},{"id":"openai/gpt-4o-mini-search-preview","object":"model","created":1755815280,"released":1741737600,"owned_by":"openai","name":"GPT 4o Mini Search Preview","description":"GPT-4o mini Search Preview is a specialized model trained to understand and execute web search queries with the Chat Completions API. In addition to token fees, web search queries have a fee per tool call.","context_window":128000,"max_tokens":16384,"type":"language","tags":["web-search"],"temperature":true,"pricing":{"input":"0.0000001575","output":"0.00000063","web_search":"10.5"}},{"id":"openai/gpt-5.2","object":"model","created":1755815280,"released":1765411200,"owned_by":"openai","name":"GPT 5.2","description":"GPT-5.2 is OpenAI's best general-purpose model, part of the GPT-5 flagship model family. It's their most intelligent model yet for both general and agentic tasks.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000018375","input":"0.0000018375","output":"0.0000147","web_search":"10.5"}},{"id":"openai/gpt-5.3-codex","object":"model","created":1755815280,"released":1770249600,"owned_by":"openai","name":"GPT 5.3 Codex","description":"GPT-5.3-Codex advances both the frontier coding performance of GPT‑5.2-Codex and the reasoning and professional knowledge capabilities of GPT‑5.2, together in one model, which is also 25% faster. This enables it to take on long-running tasks that involve research, tool use, and complex execution.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000018375","input":"0.0000018375","output":"0.0000147","web_search":"10.5"}},{"id":"openai/gpt-4o","object":"model","created":1755815280,"released":1715558400,"owned_by":"openai","name":"GPT-4o","description":"GPT-4o from OpenAI has broad general knowledge and domain expertise allowing it to follow complex instructions in natural language and solve difficult problems accurately. It matches GPT-4 Turbo performance with a faster and cheaper API.","context_window":128000,"max_tokens":16384,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.0000013125","input":"0.000002625","output":"0.0000105","web_search":"10.5"}},{"id":"meta/llama-4-maverick","object":"model","created":1755815280,"released":1743811200,"owned_by":"meta","name":"Llama 4 Maverick 17B Instruct","description":"As a general purpose LLM, Llama 4 Maverick contains 17 billion active parameters, 128 experts, and 400 billion total parameters, offering high quality at a lower price compared to Llama 3.3 70B.","context_window":128000,"max_tokens":8192,"type":"language","tags":["tool-use","vision"],"temperature":true,"pricing":{"input":"0.000000252","output":"0.0000010185"}},{"id":"openai/gpt-4o-fast","object":"model","created":1755815280,"released":1715558400,"owned_by":"openai","name":"GPT-4o (Fast)","description":"GPT-4o from OpenAI has broad general knowledge and domain expertise allowing it to follow complex instructions in natural language and solve difficult problems accurately. It matches GPT-4 Turbo performance with a faster and cheaper API.","context_window":128000,"max_tokens":16384,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.00000223125","input":"0.0000044625","output":"0.00001785","web_search":"10.5"}},{"id":"openai/gpt-4o-mini-fast","object":"model","created":1755815280,"released":1721260800,"owned_by":"openai","name":"GPT-4o mini (Fast)","description":"GPT-4o mini from OpenAI is their most advanced and cost-efficient small model. It is multi-modal (accepting text or image inputs and outputting text) and has higher intelligence than gpt-3.5-turbo but is just as fast.","context_window":128000,"max_tokens":16384,"type":"language","tags":["file-input","implicit-caching","tool-use","vision","web-search","fast"],"temperature":true,"pricing":{"input_cache_read":"0.00000013125","input":"0.0000002625","output":"0.00000105","web_search":"10.5"}},{"id":"openai/gpt-5","object":"model","created":1755815280,"released":1754524800,"owned_by":"openai","name":"GPT-5","description":"GPT-5 is OpenAI's flagship language model that excels at complex reasoning, broad real-world knowledge, code-intensive, and multi-step agentic tasks.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.00000013125","input":"0.0000013125","output":"0.0000105","web_search":"10.5"}},{"id":"openai/o3-deep-research","object":"model","created":1755815280,"released":1750896000,"owned_by":"openai","name":"o3-deep-research","description":"o3-deep-research is OpenAI's most advanced model for deep research, designed to tackle complex, multi-step research tasks. It can search and synthesize information from across the internet as well as from your own data—brought in through MCP connectors.","context_window":200000,"max_tokens":100000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["medium"]}],"temperature":true,"pricing":{"input_cache_read":"0.000002625","input":"0.0000105","output":"0.000042","web_search":"10.5"}},{"id":"openai/gpt-5.4-pro","object":"model","created":1755815280,"released":1772668800,"owned_by":"openai","name":"GPT 5.4 Pro","description":"GPT-5.4 Pro uses more compute to think harder and provide consistently better answers. It's designed to tackle tough problems.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["file-input","reasoning","tool-use","vision","web-search"],"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"temperature":true,"pricing":{"input":"0.0000315","output":"0.000189","web_search":"10.5"}},{"id":"deepseek/deepseek-v3.2-thinking","object":"model","created":1755815280,"released":1764547200,"owned_by":"deepseek","name":"DeepSeek V3.2 Thinking","description":"DeepSeek‑V3.2 from DeepSeek harmonizes high computational efficiency with superior reasoning and agent performance. It builds on three main techniques: DeepSeek Sparse Attention for long‑context efficiency, a scalable reinforcement learning framework, and a large‑scale agentic task synthesis pipeline. This model excels at long-context reasoning and agentic tasks, efficiently handling extended inputs while maintaining strong accuracy. Its sparse attention design enables it to process complex, multi-step workflows without excessive compute costs. Overall, DeepSeek‑V3.2 targets long‑context reasoning, tool‑using agents, and efficient deployment in production environments.","context_window":128000,"max_tokens":8000,"type":"language","tags":["tool-use","implicit-caching"],"temperature":true,"pricing":{"input":"0.000000651","output":"0.0000019425"}},{"id":"meta/muse-spark-1.1","object":"model","created":1755815280,"released":1783555200,"owned_by":"meta","name":"Muse Spark 1.1","description":"Muse Spark 1.1 is strongest at agentic performance, tool use, and computer use. It does well on long-running tasks with 1M token context window, can delegate execution to sub-agents running in parallel, and is trained to use computer interfaces on desktop, mobile, or browser.","context_window":1048576,"max_tokens":1048576,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision"],"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000001575","input":"0.0000013125","output":"0.0000044625"}},{"id":"openai/o4-mini-fast","object":"model","created":1755815280,"released":1744761600,"owned_by":"openai","name":"o4-mini (Fast)","description":"OpenAI's o4-mini delivers fast, cost-efficient reasoning with exceptional performance for its size, particularly excelling in math (best-performing on AIME benchmarks), coding, and visual tasks.","context_window":200000,"max_tokens":100000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.000000525","input":"0.0000021","output":"0.0000084","web_search":"10.5"}},{"id":"sakana/fugu-ultra","object":"model","created":1755815280,"released":1782000000,"owned_by":"sakana","name":"Fugu Ultra","description":"Fugu Ultra coordinates a deeper pool of expert agents to maximize answer quality on hard, high-stakes problems. Fugu Ultra focuses on maximizing performance, for a higher cost. It routes between one to three agents, depending on the problem.","context_window":1000000,"max_tokens":1000000,"type":"language","tags":["vision","tool-use","reasoning"],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input":"0.00000525","output":"0.0000315"}},{"id":"minimax/minimax-m3","object":"model","created":1755815280,"released":1780185600,"owned_by":"minimax","name":"MiniMax M3","description":"MiniMax-M3 is a frontier-class foundation model that unites the three capabilities defining today's frontier: a 1M-token context window, frontier coding and agentic performance, and native multimodality — the first open-weight model to deliver all three in a single system.","context_window":1000000,"max_tokens":1000000,"type":"language","tags":["reasoning","tool-use","vision","implicit-caching","file-input"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000063","input":"0.000000315","output":"0.00000126"}},{"id":"mistral/codestral","object":"model","created":1755815280,"released":1716940800,"owned_by":"mistral","name":"Mistral Codestral","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.","context_window":128000,"max_tokens":4000,"type":"language","tags":["tool-use"],"temperature":true,"pricing":{"input":"0.000000315","output":"0.000000945"}},{"id":"openai/gpt-5.4-mini","object":"model","created":1755815280,"released":1773705600,"owned_by":"openai","name":"GPT 5.4 Mini","description":"GPT-5.4 Mini brings the strengths of GPT-5.4 to a faster, more efficient model designed for high-volume workloads.","context_window":400000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000007875","input":"0.0000007875","output":"0.000004725","web_search":"10.5"}},{"id":"spacexai/grok-4.20-multi-agent","object":"model","created":1755815280,"released":1773100800,"owned_by":"spacexai","name":"Grok 4.20 Multi-Agent","description":"Multiple agents collaborate in parallel to perform deep research tasks.","context_window":2000000,"max_tokens":2000000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision","file-input","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000013125","output":"0.000002625","web_search":"5.25"}},{"id":"spacexai/grok-4.20-multi-agent-beta","object":"model","created":1755815280,"released":1773187200,"owned_by":"spacexai","name":"Grok 4.20 Multi Agent Beta","description":"Multiple agents collaborate in parallel to perform deep research tasks.","context_window":2000000,"max_tokens":2000000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision","file-input","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000013125","output":"0.000002625","web_search":"5.25"}},{"id":"meta/muse-spark-1.2","object":"model","created":1755815280,"released":1785888000,"owned_by":"meta","name":"Muse Spark 1.2","description":"A coding-optimized model purpose-built for agentic workflows. Improvements to code generation, debugging, and codebase understanding — with a 1M context window that handles your entire project in one session.","context_window":1048576,"max_tokens":1048576,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision"],"temperature":true,"pricing":{"input_cache_read":"0.0000001575","input":"0.0000013125","output":"0.0000044625"}},{"id":"meta/muse-spark-1.2-contributor","object":"model","created":1755815280,"released":1785888000,"owned_by":"meta","name":"Muse Spark 1.2 Contributor","description":"A coding-optimized model with pricing designed for builders. Same model, same capabilities — up to 95% less than Standard. Your inputs and outputs are used to train and improve Meta's AI models.","context_window":1048576,"max_tokens":1048576,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision"],"temperature":true,"pricing":{"input_cache_read":"0.0000000021","input":"0.000000105","output":"0.00000021"}},{"id":"openai/gpt-5.6-luna-fast","object":"model","created":1755815280,"released":1783555200,"owned_by":"openai","name":"GPT 5.6 Luna (Fast)","description":"GPT-5.6 Luna is a fast, affordable GPT-5.6 model that brings strong capability at the lowest cost in the series.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["reasoning","file-input","tool-use","vision","implicit-caching","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000042","input_cache_write":"0.0000002625","input":"0.00000042","output":"0.00000252","web_search":"10.5"}},{"id":"openai/gpt-5.6-sol","object":"model","created":1755815280,"released":1783555200,"owned_by":"openai","name":"GPT 5.6 Sol","description":"GPT-5.6 Sol is the flagship of OpenAI's GPT-5.6 series, its most capable model for long-horizon agentic work across coding, biology, and cybersecurity.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input_cache_write":"0.000002625","input":"0.0000021","output":"0.0000105","web_search":"10.5"}},{"id":"openai/gpt-5.6-terra","object":"model","created":1755815280,"released":1783555200,"owned_by":"openai","name":"GPT 5.6 Terra","description":"GPT-5.6 Terra is a balanced GPT-5.6 model for everyday work, with performance comparable to the previous generation at half the cost.","context_window":1050000,"max_tokens":128000,"type":"language","tags":["reasoning","web-search","file-input","tool-use","vision","implicit-caching","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input_cache_write":"0.000002625","input":"0.0000021","output":"0.0000126","web_search":"10.5"}},{"id":"openai/o3-mini","object":"model","created":1755815280,"released":1738281600,"owned_by":"openai","name":"o3-mini","description":"o3-mini is OpenAI's most recent small reasoning model, providing high intelligence at the same cost and latency targets of o1-mini.","context_window":200000,"max_tokens":100000,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.0000005775","input":"0.000001155","output":"0.00000462"}},{"id":"openai/o4-mini","object":"model","created":1755815280,"released":1744761600,"owned_by":"openai","name":"o4-mini","description":"OpenAI's o4-mini delivers fast, cost-efficient reasoning with exceptional performance for its size, particularly excelling in math (best-performing on AIME benchmarks), coding, and visual tasks.","context_window":200000,"max_tokens":100000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","fast"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":false,"pricing":{"input_cache_read":"0.00000028875","input":"0.000001155","output":"0.00000462","web_search":"10.5"}},{"id":"poolside/laguna-s-2.1","object":"model","created":1755815280,"released":1784505600,"owned_by":"poolside","name":"Laguna S 2.1","description":"Laguna S 2.1 is Poolside's new open-weight model for agentic coding and long-horizon work.","context_window":1000000,"max_tokens":131072,"type":"language","tags":["reasoning","tool-use"],"temperature":true,"pricing":{"input_cache_read":"0.0000000105","input":"0.000000105","output":"0.00000021"}},{"id":"poolside/laguna-s-2.1-free","object":"model","created":1755815280,"released":1784592000,"owned_by":"poolside","name":"Laguna S 2.1 Free","description":"Laguna S 2.1 is Poolside's new open-weight model for agentic coding and long-horizon work.","context_window":256000,"max_tokens":32768,"type":"language","tags":["reasoning","tool-use","free"],"temperature":true,"pricing":{"input":"0","output":"0"}},{"id":"stepfun/step-3.5-flash","object":"model","created":1755815280,"released":1769644800,"owned_by":"stepfun","name":"StepFun 3.5 Flash","description":"Step 3.5 Flash is an open-source reasoning model by StepFun with 196B total parameters (11B active) using Mixture of Experts. It features a 256K context window, deep reasoning, tool calling, and agentic capabilities, achieving 97.3 on AIME 2025 and 74.4% on SWE-bench Verified.","context_window":262114,"max_tokens":262114,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000021","input":"0.0000000945","output":"0.000000315"}},{"id":"thinkingmachines/inkling-small","object":"model","created":1755815280,"released":1785369600,"owned_by":"thinkingmachines","name":"Inkling Small","description":"Inkling-Small is a lighter-weight model with 12B active parameters, trained with a similar recipe, to Inkling that achieves strong performance with even lower cost and latency.","context_window":1000000,"max_tokens":1000000,"type":"language","tags":["reasoning","tool-use","vision","file-input","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000105","input":"0.000000525","output":"0.00000126"}},{"id":"spacexai/grok-4.20-reasoning-beta","object":"model","created":1755815280,"released":1773187200,"owned_by":"spacexai","name":"Grok 4.20 Beta Reasoning","description":"Grok 4.20 Beta is the newest flagship model from xAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherance, delivering consistently precise and truthful responses.","context_window":2000000,"max_tokens":2000000,"type":"language","tags":["reasoning","tool-use","vision","file-input","implicit-caching","web-search"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000013125","output":"0.000002625","web_search":"5.25"}},{"id":"spacexai/grok-4.5","object":"model","created":1755815280,"released":1783468800,"owned_by":"spacexai","name":"Grok 4.5","description":"SpaceXAI's smartest model with frontier performance on coding, knowledge work, and STEM.","context_window":500000,"max_tokens":500000,"type":"language","tags":["reasoning","tool-use","implicit-caching","file-input","vision","web-search"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000315","input":"0.0000021","output":"0.0000063","web_search":"5.25"}},{"id":"xiaomi/mimo-v2.5","object":"model","created":1755815280,"released":1776816000,"owned_by":"xiaomi","name":"MiMo M2.5","description":"A native full-modal model supporting text, image, video, and audio understanding, with powerful Agent capabilities.","context_window":1050000,"max_tokens":131100,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.00000000294","input":"0.000000147","output":"0.000000294"}},{"id":"tencent/hy-mt2-lite","object":"model","created":1755815280,"released":1781222400,"owned_by":"tencent","name":"Tencent Hy-MT2-Lite","description":"Hy-MT2-Lite is Tencent Cloud Hunyuan’s lightweight 1.8B translation model, combining efficient performance with an 8K-token context window and broad multilingual coverage. It supports translation across 33 languages and five ethnic Chinese and dialect variants, delivering strong results on benchmarks including FLORES-200 and WMT25. Designed for professional and real-world business use, it reliably follows instructions for structured, delimiter-preserving, context-aware, glossary-guided, and style-specific translation.","context_window":8000,"max_tokens":4000,"type":"language","tags":[],"temperature":true,"pricing":{"input":"0.0000000462","output":"0.00000018585"}},{"id":"spacexai/grok-4.20-non-reasoning","object":"model","created":1755815280,"released":1773100800,"owned_by":"spacexai","name":"Grok 4.20 Non-Reasoning","description":"Grok 4.20 Beta is the newest flagship model from xAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherence, delivering consistently precise and truthful responses.","context_window":2000000,"max_tokens":2000000,"type":"language","tags":["tool-use","implicit-caching","file-input","vision","web-search"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000013125","output":"0.000002625","web_search":"5.25"}},{"id":"spacexai/grok-4.20-non-reasoning-beta","object":"model","created":1755815280,"released":1773187200,"owned_by":"spacexai","name":"Grok 4.20 Beta Non-Reasoning","description":"Grok 4.20 Beta is the newest flagship model from xAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherance, delivering consistently precise and truthful responses.","context_window":2000000,"max_tokens":2000000,"type":"language","tags":["tool-use","implicit-caching","vision","file-input","web-search"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000013125","output":"0.000002625","web_search":"5.25"}},{"id":"spacexai/grok-4.6","object":"model","created":1755815280,"released":1786492800,"owned_by":"spacexai","name":"Grok 4.6","description":"Grok 4.6 builds on Grok 4.5 with a particular focus on long-running agents and more ambitious interactive and visual work. It stays with complex tasks across many steps, whether researching a topic, analyzing information, working across a codebase, or turning an idea into a polished application or work artifact.","context_window":500000,"max_tokens":500000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision"],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input":"0.0000021","output":"0.0000063","web_search":"5.25"}},{"id":"stepfun/step-3.7-flash","object":"model","created":1755815280,"released":1779926400,"owned_by":"stepfun","name":"Step 3.7 Flash","description":"StepFun’s flagship multimodal reasoning model. Powered by a 198B-parameter / 11B-activation sparse MoE architecture, with native support for image and video understanding.","context_window":256000,"max_tokens":256000,"type":"language","tags":["implicit-caching","reasoning","tool-use","vision"],"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000042","input":"0.00000021","output":"0.0000012075"}},{"id":"openai/gpt-5.5-fast","object":"model","created":1755815280,"released":1776988800,"owned_by":"openai","name":"GPT 5.5 (Fast)","description":"GPT‑5.5 understands what you’re trying to do faster and can carry more of the work itself. It excels at writing and debugging code, researching online, analyzing data, creating documents and spreadsheets, operating software, and moving across tools until a task is finished. Instead of carefully managing every step, you can give GPT‑5.5 a messy, multi-part task and trust it to plan, use tools, check its work, navigate through ambiguity, and keep going.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000013125","input":"0.000013125","output":"0.00007875","web_search":"10.5"}},{"id":"tencent/hy-mt2-plus","object":"model","created":1755815280,"released":1781222400,"owned_by":"tencent","name":"Tencent Hy-MT2-Plus","description":"Hy-MT2-Plus is Tencent Cloud Hunyuan’s 7B translation model, offering an 8K-token context window and broad multilingual coverage. Optimized for translation across 33 languages and five ethnic Chinese and dialect variants, it delivers industry-leading performance on benchmarks including FLORES-200 and WMT25. It excels in professional domains and real-world business scenarios, with strong instruction-following for structured, delimiter-preserving, context-aware, glossary-guided, and style-specific translation.","context_window":8000,"max_tokens":4000,"type":"language","tags":[],"temperature":true,"pricing":{"input":"0.0000000777","output":"0.00000030975"}},{"id":"spacexai/grok-build-0.1","object":"model","created":1755815280,"released":1779235200,"owned_by":"spacexai","name":"Grok Build 0.1","description":"xAI's fast coding model trained specifically for agentic coding.","context_window":256000,"max_tokens":256000,"type":"language","tags":["reasoning","implicit-caching","vision","tool-use","web-search"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.00000105","output":"0.0000021","web_search":"5.25"}},{"id":"zai/glm-4.7","object":"model","created":1755815280,"released":1766361600,"owned_by":"zai","name":"GLM 4.7","description":"GLM-4.7 is Z.ai’s latest flagship model, with major upgrades focused on two key areas: stronger coding capabilities and more stable multi-step reasoning and execution.","context_window":200000,"max_tokens":120000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000126","input":"0.00000063","output":"0.00000231"}},{"id":"openai/gpt-5.5","object":"model","created":1755815280,"released":1776988800,"owned_by":"openai","name":"GPT 5.5","description":"GPT‑5.5 understands what you’re trying to do faster and can carry more of the work itself. It excels at writing and debugging code, researching online, analyzing data, creating documents and spreadsheets, operating software, and moving across tools until a task is finished. Instead of carefully managing every step, you can give GPT‑5.5 a messy, multi-part task and trust it to plan, use tools, check its work, navigate through ambiguity, and keep going.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["file-input","implicit-caching","reasoning","tool-use","vision","web-search","websocket-realtime","fast"],"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000525","input":"0.00000525","output":"0.0000315","web_search":"10.5"}},{"id":"zai/glm-4.7-flash","object":"model","created":1755815280,"released":1768780800,"owned_by":"zai","name":"GLM 4.7 Flash","description":"GLM-4.7-Flash balances high performance with efficiency, making it the perfect lightweight deployment option. Beyond coding, it is also recommended for creative writing, translation, long-context tasks, and roleplay.","context_window":200000,"max_tokens":131000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.0000000735","output":"0.00000042"}},{"id":"tencent/hy-mt2-pro","object":"model","created":1755815280,"released":1779321600,"owned_by":"tencent","name":"Tencent Hy-MT2-Pro","description":"Hy-MT2-Pro is Tencent Cloud Hunyuan’s flagship translation model, featuring 30B-A3B parameters and an 8K-token context window. It delivers bidirectional translation across 33 core language pairs, plus five minority-language and dialect pairs. With industry-leading performance on FLORES-200 and WMT25, it excels in specialized domains and real-world business scenarios while supporting structured, contextual, terminology-aware, style-specific, and delimiter-preserving translation.","context_window":8000,"max_tokens":4000,"type":"language","tags":[],"temperature":true,"pricing":{"input":"0.0000000777","output":"0.00000030975"}},{"id":"thinkingmachines/inkling","object":"model","created":1755815280,"released":1784073600,"owned_by":"thinkingmachines","name":"Inkling","description":"Inkling is a multimodal MoE model (975B total, 41B active, 256k context) reasoning over text, image, and audio inputs.","context_window":256000,"max_tokens":256000,"type":"language","tags":["reasoning","tool-use","vision","file-input"],"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000001785","input":"0.00000105","output":"0.0000042525"}},{"id":"zai/glm-4.7-flashx","object":"model","created":1755815280,"released":1768780800,"owned_by":"zai","name":"GLM 4.7 FlashX","description":"GLM-4.7-Flash balances high performance with efficiency, making it the perfect lightweight deployment option.","context_window":200000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.0000000105","input":"0.000000063","output":"0.00000042"}},{"id":"zai/glm-5-turbo","object":"model","created":1755815280,"released":1773532800,"owned_by":"zai","name":"GLM 5 Turbo","description":"GLM 5 Turbo is a foundation model deeply optimized for the OpenClaw scenario. It has been specifically optimized for the core requirements of OpenClaw tasks since the training phase, enhancing key capabilities such as tool invocation, command following, timed and persistent tasks, and long-chain execution.","context_window":202800,"max_tokens":131100,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000252","input":"0.00000126","output":"0.0000042"}},{"id":"zai/glm-5.1","object":"model","created":1755815280,"released":1775520000,"owned_by":"zai","name":"GLM 5.1","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 a single task for more than 8 hours—autonomously planning, executing, and improving itself throughout the process—ultimately delivering complete, engineering-grade results.","context_window":202800,"max_tokens":64000,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000273","input":"0.00000147","output":"0.00000462"}},{"id":"zai/glm-5.3","object":"model","created":1755815280,"released":1787011200,"owned_by":"zai","name":"GLM 5.3","description":"GLM 5.3 delivers comprehensive advancements in complex software engineering and agent capabilities. It uses the same base model as GLM-5.2, with all improvements driven by post-training.","context_window":1000000,"max_tokens":12800,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"temperature":true,"pricing":{"input_cache_read":"0.000000273","input":"0.00000147","output":"0.00000462"}},{"id":"zai/glm-4.5-air","object":"model","created":1755815280,"released":1753660800,"owned_by":"zai","name":"GLM 4.5 Air","description":"GLM-4.5 and GLM-4.5-Air are our latest flagship models, purpose-built as foundational models for agent-oriented applications. Both leverage a Mixture-of-Experts (MoE) architecture. GLM-4.5 has a total parameter count of 355B with 32B active parameters per forward pass, while GLM-4.5-Air adopts a more streamlined design with 106B total parameters and 12B active parameters.","context_window":128000,"max_tokens":96000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.0000000315","input":"0.00000021","output":"0.000001155"}},{"id":"zai/glm-4.5v","object":"model","created":1755815280,"released":1754870400,"owned_by":"zai","name":"GLM 4.5V","description":"Built on the GLM-4.5-Air base model, GLM-4.5V inherits proven techniques from GLM-4.1V-Thinking while achieving effective scaling through a powerful 106B-parameter MoE architecture.","context_window":66000,"max_tokens":16000,"type":"language","tags":["implicit-caching","reasoning","tool-use","vision"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.0000001155","input":"0.00000063","output":"0.00000189"}},{"id":"zai/glm-4.6","object":"model","created":1755815280,"released":1759190400,"owned_by":"zai","name":"GLM 4.6","description":"As the latest iteration in the GLM series, GLM-4.6 achieves comprehensive enhancements across multiple domains, including real-world coding, long-context processing, reasoning, searching, writing, and agentic applications.","context_window":200000,"max_tokens":96000,"type":"language","tags":["implicit-caching","reasoning","tool-use"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.0000001155","input":"0.00000063","output":"0.00000231"}},{"id":"spacexai/grok-4.20-reasoning","object":"model","created":1755815280,"released":1773100800,"owned_by":"spacexai","name":"Grok 4.20 Reasoning","description":"Grok 4.20 Beta is the newest flagship model from xAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherence, delivering consistently precise and truthful responses.","context_window":2000000,"max_tokens":2000000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision","file-input","web-search"],"temperature":true,"pricing":{"input_cache_read":"0.00000021","input":"0.0000013125","output":"0.000002625","web_search":"5.25"}},{"id":"zai/glm-5.2","object":"model","created":1755815280,"released":1781568000,"owned_by":"zai","name":"GLM 5.2","description":"GLM-5.2 delivers powerful coding capabilities, usable 1M-context support, and continued strengths in long-horizon tasks.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.000000168","input":"0.00000084","output":"0.0000026775"}},{"id":"zai/glm-5v-turbo","object":"model","created":1755815280,"released":1775001600,"owned_by":"zai","name":"GLM 5V Turbo","description":"GLM-5V-Turbo is Z.AI’s first multimodal coding foundation model, built for vision-based coding tasks. It can natively process multimodal inputs such as images, video, and text, while also excelling at long-horizon planning, complex coding, and action execution. Deeply optimized for agent workflows, it works seamlessly with agents such as Claude Code and OpenClaw to complete the full loop of “understand the environment → plan actions → execute tasks”.","context_window":200000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","implicit-caching","vision","file-input"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.000000252","input":"0.00000126","output":"0.0000042"}},{"id":"zai/glm-5","object":"model","created":1755815280,"released":1770854400,"owned_by":"zai","name":"GLM 5","description":"GLM 5 is a frontier-class, general-purpose large language model optimized for complex systems engineering and long-horizon agentic tasks. It builds on the GLM 4.5 agent-centric lineage and is designed to support multi-step reasoning, math (including AIME-style benchmarks), advanced coding, and tool-augmented workflows, with long context support suitable for sophisticated agents and enterprise applications. Typical uses include autonomous agents for software engineering, data and systems troubleshooting, operations copilots, and high-end chat assistants that must break down complex tasks, call tools reliably, and reason over long sequences of instructions or documents.","context_window":202800,"max_tokens":131100,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input":"0.00000105","output":"0.00000336"}},{"id":"zai/glm-5.2-fast","object":"model","created":1755815280,"released":1782172800,"owned_by":"zai","name":"GLM 5.2 Fast","description":"Fast version of GLM 5.2 with 120-250 TPS.","context_window":1000000,"max_tokens":128000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","xhigh"]}],"temperature":true,"pricing":{"input_cache_read":"0.0000002205","input":"0.000002205","output":"0.00000693"}},{"id":"xiaomi/mimo-v2.5-pro","object":"model","created":1755815280,"released":1776816000,"owned_by":"xiaomi","name":"MiMo V2.5 Pro","description":"MiMo V2.5 Pro delivers significant improvements over its predecessor, MiMo-V2-Pro, in general agentic capabilities, complex software engineering, and long-horizon tasks. MiMo-V2.5-Pro is a 1.02T-parameter Mixture-of-Experts model with 42B active parameters, built on a hybrid-attention architecture with a 1M-token context window.","context_window":1050000,"max_tokens":131000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.00000000378","input":"0.00000045675","output":"0.0000009135"}},{"id":"zai/glm-4.5","object":"model","created":1755815280,"released":1753660800,"owned_by":"zai","name":"GLM 4.5","description":"GLM-4.5 and GLM-4.5-Air are our latest flagship models, purpose-built as foundational models for agent-oriented applications. Both leverage a Mixture-of-Experts (MoE) architecture. GLM-4.5 has a total parameter count of 355B with 32B active parameters per forward pass, while GLM-4.5-Air adopts a more streamlined design with 106B total parameters and 12B active parameters.","context_window":128000,"max_tokens":96000,"type":"language","tags":["reasoning","tool-use","implicit-caching"],"reasoning_options":[{"type":"toggle"}],"temperature":true,"pricing":{"input_cache_read":"0.0000001155","input":"0.00000063","output":"0.00000231"}}]}