{"data":[{"id":"nvidia/nemotron-nano-9b-v2:free","canonical_slug":"nvidia/nemotron-nano-9b-v2","hugging_face_id":"nvidia/NVIDIA-Nemotron-Nano-9B-v2","name":"NVIDIA: Nemotron Nano 9B V2 (free)","created":1757106807,"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. \n\nThe 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_length":32000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0","completion":"0"},"top_provider":{"context_length":128000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen3-next-80b-a3b-instruct:free","canonical_slug":"qwen/qwen3-next-80b-a3b-instruct-2509","hugging_face_id":"Qwen/Qwen3-Next-80B-A3B-Instruct","name":"Qwen: Qwen3 Next 80B A3B Instruct (free)","created":1757612213,"description":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks across reasoning, code generation, knowledge QA, and multilingual use, while remaining robust on alignment and formatting. Compared with prior Qwen3 instruct variants, it focuses on higher throughput and stability on ultra-long inputs and multi-turn dialogues, making it well-suited for RAG, tool use, and agentic workflows that require consistent final answers rather than visible chain-of-thought.\n\nThe model employs scaling-efficient training and decoding to improve parameter efficiency and inference speed, and has been validated on a broad set of public benchmarks where it reaches or approaches larger Qwen3 systems in several categories while outperforming earlier mid-sized baselines. It is best used as a general assistant, code helper, and long-context task solver in production settings where deterministic, instruction-following outputs are preferred.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0","completion":"0"},"top_provider":{"context_length":262144,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-09-30","expiration_date":null},{"id":"cognitivecomputations/dolphin-mistral-24b-venice-edition:free","canonical_slug":"venice/uncensored","hugging_face_id":"cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition","name":"Venice: Uncensored (free)","created":1752094966,"description":"Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai. This model is designed as an “uncensored” instruct-tuned LLM, preserving user control over alignment, system prompts, and behavior. Intended for advanced and unrestricted use cases, Venice Uncensored emphasizes steerability and transparent behavior, removing default safety and alignment layers typically found in mainstream assistant models.","context_length":32768,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0","completion":"0"},"top_provider":{"context_length":32768,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-04-30","expiration_date":null},{"id":"arcee-ai/trinity-mini:free","canonical_slug":"arcee-ai/trinity-mini-20251201","hugging_face_id":"arcee-ai/Trinity-Mini","name":"Arcee AI: Trinity Mini (free)","created":1764601720,"description":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token. Engineered for efficient reasoning over long contexts (131k) with robust function calling and multi-step agent workflows.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0","completion":"0"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.15,"top_p":0.75,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":"2026-04-10"},{"id":"qwen/qwen3.6-plus:free","canonical_slug":"qwen/qwen3.6-plus-04-02","hugging_face_id":"","name":"Qwen: Qwen3.6 Plus (free)","created":1775133557,"description":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference. Compared to the 3.5 series, it delivers major gains in agentic coding, front-end development, and overall reasoning, with a significantly improved “vibe coding” experience. The model excels at complex tasks such as 3D scenes, games, and repository-level problem solving, achieving a 78.8 score on SWE-bench Verified. It represents a substantial leap in both pure-text and multimodal capabilities, performing at the level of leading state-of-the-art models.","context_length":1000000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0","completion":"0"},"top_provider":{"context_length":1000000,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","presence_penalty","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"arcee-ai/trinity-large-preview:free","canonical_slug":"arcee-ai/trinity-large-preview","hugging_face_id":"arcee-ai/Trinity-Large-Preview","name":"Arcee AI: Trinity Large Preview (free)","created":1769552670,"description":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing. \n\nIt excels in creative writing, storytelling, role-play, chat scenarios, and real-time voice assistance, better than your average reasoning model usually can. But we’re also introducing some of our newer agentic performance. It was trained to navigate well in agent harnesses like OpenCode, Cline, and Kilo Code, and to handle complex toolchains and long, constraint-filled prompts. \n\nThe architecture natively supports very long context windows up to 512k tokens, with the Preview API currently served at 128k context using 8-bit quantization for practical deployment. Trinity-Large-Preview reflects Arcee’s efficiency-first design philosophy, offering a production-oriented frontier model with open weights and permissive licensing suitable for real-world applications and experimentation.","context_length":131000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0","completion":"0"},"top_provider":{"context_length":131000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","structured_outputs","temperature","tools","top_k","top_p"],"default_parameters":{"temperature":0.8,"top_p":0.8,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"nvidia/nemotron-3-super-120b-a12b:free","canonical_slug":"nvidia/nemotron-3-super-120b-a12b-20230311","hugging_face_id":"nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8","name":"NVIDIA: Nemotron 3 Super (free)","created":1773245239,"description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer Mixture-of-Experts architecture with multi-token prediction (MTP), it delivers over 50% higher token generation compared to leading open models.\n \nThe model features a 1M token context window for long-term agent coherence, cross-document reasoning, and multi-step task planning. Latent MoE enables calling 4 experts for the inference cost of only one, improving intelligence and generalization. Multi-environment RL training across 10+ environments delivers leading accuracy on benchmarks including AIME 2025, TerminalBench, and SWE-Bench Verified.\n \nFully open with weights, datasets, and recipes under the NVIDIA Open License, Nemotron 3 Super allows easy customization and secure deployment anywhere — from workstation to cloud.","context_length":1000000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0","completion":"0"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"meta-llama/llama-3.1-8b-instruct","canonical_slug":"meta-llama/llama-3.1-8b-instruct","hugging_face_id":"meta-llama/Meta-Llama-3.1-8B-Instruct","name":"Meta: Llama 3.1 8B Instruct","created":1721692800,"description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3-1/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/).","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"llama3"},"pricing":{"prompt":"0.00000002","completion":"0.00000005"},"top_provider":{"context_length":131072,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"qwen/qwen2.5-coder-7b-instruct","canonical_slug":"qwen/qwen2.5-coder-7b-instruct","hugging_face_id":"Qwen/Qwen2.5-Coder-7B-Instruct","name":"Qwen: Qwen2.5 Coder 7B Instruct","created":1744734887,"description":"Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing. Based on the Qwen2.5 architecture, it incorporates enhancements like RoPE, SwiGLU, RMSNorm, and GQA attention with support for up to 128K tokens using YaRN-based extrapolation. It is trained on a large corpus of source code, synthetic data, and text-code grounding, providing robust performance across programming languages and agentic coding workflows.\n\nThis model is part of the Qwen2.5-Coder family and offers strong compatibility with tools like vLLM for efficient deployment. Released under the Apache 2.0 license.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen","instruct_type":null},"pricing":{"prompt":"0.00000003","completion":"0.00000009"},"top_provider":{"context_length":32768,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","repetition_penalty","response_format","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"openai/gpt-oss-20b","canonical_slug":"openai/gpt-oss-20b","hugging_face_id":"openai/gpt-oss-20b","name":"OpenAI: gpt-oss-20b","created":1754414229,"description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference and deployability on consumer or single-GPU hardware. The model is trained in OpenAI’s Harmony response format and supports reasoning level configuration, fine-tuning, and agentic capabilities including function calling, tool use, and structured outputs.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000003","completion":"0.00000011","input_cache_read":"0.000000015"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","reasoning_effort","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"mistralai/mistral-small-3.1-24b-instruct","canonical_slug":"mistralai/mistral-small-3.1-24b-instruct-2503","hugging_face_id":"mistralai/Mistral-Small-3.1-24B-Instruct-2503","name":"Mistral: Mistral Small 3.1 24B","created":1742238937,"description":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities. It provides state-of-the-art performance in text-based reasoning and vision tasks, including image analysis, programming, mathematical reasoning, and multilingual support across dozens of languages. Equipped with an extensive 128k token context window and optimized for efficient local inference, it supports use cases such as conversational agents, function calling, long-document comprehension, and privacy-sensitive deployments. The updated version is [Mistral Small 3.2](mistralai/mistral-small-3.2-24b-instruct)","context_length":128000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.00000003","completion":"0.00000011","input_cache_read":"0.000000015"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"cohere/command-r7b-12-2024","canonical_slug":"cohere/command-r7b-12-2024","hugging_face_id":"","name":"Cohere: Command R7B (12-2024)","created":1734158152,"description":"Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG, tool use, agents, and similar tasks requiring complex reasoning and multiple steps.\n\nUse of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Cohere","instruct_type":null},"pricing":{"prompt":"0.0000000375","completion":"0.00000015"},"top_provider":{"context_length":128000,"max_completion_tokens":4000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-08-31","expiration_date":null},{"id":"mistralai/mistral-nemo","canonical_slug":"mistralai/mistral-nemo","hugging_face_id":"mistralai/Mistral-Nemo-Instruct-2407","name":"Mistral: Mistral Nemo","created":1721347200,"description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.\n\nThe model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, and Hindi.\n\nIt supports function calling and is released under the Apache 2.0 license.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":"mistral"},"pricing":{"prompt":"0.00000004","completion":"0.00000017"},"top_provider":{"context_length":60288,"max_completion_tokens":16000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2024-04-30","expiration_date":null},{"id":"arcee-ai/trinity-mini","canonical_slug":"arcee-ai/trinity-mini-20251201","hugging_face_id":"arcee-ai/Trinity-Mini","name":"Arcee AI: Trinity Mini","created":1764601720,"description":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token. Engineered for efficient reasoning over long contexts (131k) with robust function calling and multi-step agent workflows.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.000000045","completion":"0.00000015"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.15,"top_p":0.75,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"sao10k/l3-lunaris-8b","canonical_slug":"sao10k/l3-lunaris-8b","hugging_face_id":"Sao10K/L3-8B-Lunaris-v1","name":"Sao10K: Llama 3 8B Lunaris","created":1723507200,"description":"Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3. It's a strategic merge of multiple models, designed to balance creativity with improved logic and general knowledge.\n\nCreated by [Sao10k](https://huggingface.co/Sao10k), this model aims to offer an improved experience over Stheno v3.2, with enhanced creativity and logical reasoning.\n\nFor best results, use with Llama 3 Instruct context template, temperature 1.4, and min_p 0.1.","context_length":8192,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"llama3"},"pricing":{"prompt":"0.00000005","completion":"0.00000005"},"top_provider":{"context_length":8192,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"qwen/qwen3.5-9b","canonical_slug":"qwen/qwen3.5-9b-20260310","hugging_face_id":"Qwen/Qwen3.5-9B","name":"Qwen: Qwen3.5-9B","created":1773152396,"description":"Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture. It uses a unified vision-language design with early fusion of multimodal tokens, allowing the model to process and reason across text and images within the same context.","context_length":256000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000005","completion":"0.00000015"},"top_provider":{"context_length":256000,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-oss-120b","canonical_slug":"openai/gpt-oss-120b","hugging_face_id":"openai/gpt-oss-120b","name":"OpenAI: gpt-oss-120b","created":1754414231,"description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000005","completion":"0.00000025"},"top_provider":{"context_length":131072,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","reasoning_effort","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"openai/gpt-5-nano","canonical_slug":"openai/gpt-5-nano-2025-08-07","hugging_face_id":"","name":"OpenAI: GPT-5 Nano","created":1754587402,"description":"GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments. While limited in reasoning depth compared to its larger counterparts, it retains key instruction-following and safety features. It is the successor to GPT-4.1-nano and offers a lightweight option for cost-sensitive or real-time applications.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000005","completion":"0.0000004","web_search":"0.01","input_cache_read":"0.000000005"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2024-05-31","expiration_date":null},{"id":"qwen/qwen3-8b","canonical_slug":"qwen/qwen3-8b-04-28","hugging_face_id":"Qwen/Qwen3-8B","name":"Qwen: Qwen3 8B","created":1745876632,"description":"Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue. It supports seamless switching between \"thinking\" mode for math, coding, and logical inference, and \"non-thinking\" mode for general conversation. The model is fine-tuned for instruction-following, agent integration, creative writing, and multilingual use across 100+ languages and dialects. It natively supports a 32K token context window and can extend to 131K tokens with YaRN scaling.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":"qwen3"},"pricing":{"prompt":"0.00000005","completion":"0.0000004","input_cache_read":"0.00000005"},"top_provider":{"context_length":40960,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"gryphe/mythomax-l2-13b","canonical_slug":"gryphe/mythomax-l2-13b","hugging_face_id":"Gryphe/MythoMax-L2-13b","name":"MythoMax 13B","created":1688256000,"description":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge","context_length":4096,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama2","instruct_type":"alpaca"},"pricing":{"prompt":"0.00000006","completion":"0.00000006"},"top_provider":{"context_length":4096,"max_completion_tokens":4096,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_a","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-06-30","expiration_date":null},{"id":"qwen/qwen3-14b","canonical_slug":"qwen/qwen3-14b-04-28","hugging_face_id":"Qwen/Qwen3-14B","name":"Qwen: Qwen3 14B","created":1745876478,"description":"Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It supports seamless switching between a \"thinking\" mode for tasks like math, programming, and logical inference, and a \"non-thinking\" mode for general-purpose conversation. The model is fine-tuned for instruction-following, agent tool use, creative writing, and multilingual tasks across 100+ languages and dialects. It natively handles 32K token contexts and can extend to 131K tokens using YaRN-based scaling.","context_length":131702,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":"qwen3"},"pricing":{"prompt":"0.00000006","completion":"0.00000024"},"top_provider":{"context_length":40960,"max_completion_tokens":40960,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"z-ai/glm-4.7-flash","canonical_slug":"z-ai/glm-4.7-flash-20260119","hugging_face_id":"zai-org/GLM-4.7-Flash","name":"Z.ai: GLM 4.7 Flash","created":1768833913,"description":"As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency. It is further optimized for agentic coding use cases, strengthening coding capabilities, long-horizon task planning, and tool collaboration, and has achieved leading performance among open-source models of the same size on several current public benchmark leaderboards.","context_length":200000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000006","completion":"0.0000004","input_cache_read":"0.0000000100000002"},"top_provider":{"context_length":202752,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"microsoft/phi-4","canonical_slug":"microsoft/phi-4","hugging_face_id":"microsoft/phi-4","name":"Microsoft: Phi 4","created":1736489872,"description":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. \n\nAt 14 billion parameters, it was trained on a mix of high-quality synthetic datasets, data from curated websites, and academic materials. It has undergone careful improvement to follow instructions accurately and maintain strong safety standards. It works best with English language inputs.\n\nFor more information, please see [Phi-4 Technical Report](https://arxiv.org/pdf/2412.08905)\n","context_length":16384,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.000000065","completion":"0.00000014"},"top_provider":{"context_length":16384,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"qwen/qwen3.5-flash-02-23","canonical_slug":"qwen/qwen3.5-flash-20260224","hugging_face_id":null,"name":"Qwen: Qwen3.5-Flash","created":1772053776,"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_length":1000000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.000000065","completion":"0.00000026"},"top_provider":{"context_length":1000000,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","presence_penalty","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"qwen/qwen3-coder-30b-a3b-instruct","canonical_slug":"qwen/qwen3-coder-30b-a3b-instruct","hugging_face_id":"Qwen/Qwen3-Coder-30B-A3B-Instruct","name":"Qwen: Qwen3 Coder 30B A3B Instruct","created":1753972379,"description":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the Qwen3 architecture, it supports a native context length of 256K tokens (extendable to 1M with Yarn) and performs strongly in tasks involving function calls, browser use, and structured code completion.\n\nThis model is optimized for instruction-following without “thinking mode”, and integrates well with OpenAI-compatible tool-use formats. ","context_length":0,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000007","completion":"0.00000027"},"top_provider":{"context_length":160000,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"bytedance-seed/seed-1.6-flash","canonical_slug":"bytedance-seed/seed-1.6-flash-20250625","hugging_face_id":"","name":"ByteDance Seed: Seed 1.6 Flash","created":1766505011,"description":"Seed 1.6 Flash is an ultra-fast multimodal deep thinking model by ByteDance Seed, supporting both text and visual understanding. It features a 256k context window and can generate outputs of up to 16k tokens.","context_length":262144,"architecture":{"modality":"text+image+video->text","input_modalities":["image","text","video"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.000000075","completion":"0.0000003"},"top_provider":{"context_length":262144,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-2.0-flash-lite-001","canonical_slug":"google/gemini-2.0-flash-lite-001","hugging_face_id":"","name":"Google: Gemini 2.0 Flash Lite","created":1740506212,"description":"Gemini 2.0 Flash Lite offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro 1.5](/google/gemini-pro-1.5), all at extremely economical token prices.","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.000000075","completion":"0.0000003","image":"0.000000075","audio":"0.000000075","internal_reasoning":"0.0000003"},"top_provider":{"context_length":1048576,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-08-31","expiration_date":"2026-06-01"},{"id":"qwen/qwen3-32b","canonical_slug":"qwen/qwen3-32b-04-28","hugging_face_id":"Qwen/Qwen3-32B","name":"Qwen: Qwen3 32B","created":1745875945,"description":"Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a \"thinking\" mode for tasks like math, coding, and logical inference, and a \"non-thinking\" mode for faster, general-purpose conversation. The model demonstrates strong performance in instruction-following, agent tool use, creative writing, and multilingual tasks across 100+ languages and dialects. It natively handles 32K token contexts and can extend to 131K tokens using YaRN-based scaling. ","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":"qwen3"},"pricing":{"prompt":"0.00000008","completion":"0.00000024","input_cache_read":"0.00000004"},"top_provider":{"context_length":40960,"max_completion_tokens":40960,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"google/gemma-3-12b-it","canonical_slug":"google/gemma-3-12b-it","hugging_face_id":"google/gemma-3-12b-it","name":"Google: Gemma 3 12B","created":1741902625,"description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 12B is the second largest in the family of Gemma 3 models after [Gemma 3 27B](google/gemma-3-27b-it)","context_length":131072,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":"gemma"},"pricing":{"prompt":"0.00000008","completion":"0.0000003","input_cache_read":"0.00000004"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-08-31","expiration_date":null},{"id":"qwen/qwen3-30b-a3b-thinking-2507","canonical_slug":"qwen/qwen3-30b-a3b-thinking-2507","hugging_face_id":"Qwen/Qwen3-30B-A3B-Thinking-2507","name":"Qwen: Qwen3 30B A3B Thinking 2507","created":1756399192,"description":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for “thinking mode,” where internal reasoning traces are separated from final answers.\n\nCompared to earlier Qwen3-30B releases, this version improves performance across logical reasoning, mathematics, science, coding, and multilingual benchmarks. It also demonstrates stronger instruction following, tool use, and alignment with human preferences. With higher reasoning efficiency and extended output budgets, it is best suited for advanced research, competitive problem solving, and agentic applications requiring structured long-context reasoning.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000008","completion":"0.0000004","input_cache_read":"0.00000008"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"google/gemma-3-27b-it","canonical_slug":"google/gemma-3-27b-it","hugging_face_id":"google/gemma-3-27b-it","name":"Google: Gemma 3 27B","created":1741756359,"description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 27B is Google's latest open source model, successor to [Gemma 2](google/gemma-2-27b-it)","context_length":131072,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":"gemma"},"pricing":{"prompt":"0.00000008","completion":"0.00000045"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-08-31","expiration_date":null},{"id":"qwen/qwen3-vl-8b-instruct","canonical_slug":"qwen/qwen3-vl-8b-instruct","hugging_face_id":"Qwen/Qwen3-VL-8B-Instruct","name":"Qwen: Qwen3 VL 8B Instruct","created":1760463308,"description":"Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video. It features improved multimodal fusion with Interleaved-MRoPE for long-horizon temporal reasoning, DeepStack for fine-grained visual-text alignment, and text-timestamp alignment for precise event localization.\n\nThe model supports a native 256K-token context window, extensible to 1M tokens, and handles both static and dynamic media inputs for tasks like document parsing, visual question answering, spatial reasoning, and GUI control. It achieves text understanding comparable to leading LLMs while expanding OCR coverage to 32 languages and enhancing robustness under varied visual conditions.","context_length":256000,"architecture":{"modality":"text+image->text","input_modalities":["image","text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000008","completion":"0.0000005"},"top_provider":{"context_length":131072,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.7,"top_p":0.8,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"xiaomi/mimo-v2-flash","canonical_slug":"xiaomi/mimo-v2-flash-20251210","hugging_face_id":"XiaomiMiMo/MiMo-V2-Flash","name":"Xiaomi: MiMo-V2-Flash","created":1765731308,"description":"MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi. It is a Mixture-of-Experts model with 309B total parameters and 15B active parameters, adopting hybrid attention architecture. MiMo-V2-Flash supports a hybrid-thinking toggle and a 256K context window, and excels at reasoning, coding, and agent scenarios. On SWE-bench Verified and SWE-bench Multilingual, MiMo-V2-Flash ranks as the top #1 open-source model globally, delivering performance comparable to Claude Sonnet 4.5 while costing only about 3.5% as much.\n\nUsers can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config).","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000009","completion":"0.00000029","input_cache_read":"0.000000045"},"top_provider":{"context_length":262144,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"qwen/qwen3-30b-a3b-instruct-2507","canonical_slug":"qwen/qwen3-30b-a3b-instruct-2507","hugging_face_id":"Qwen/Qwen3-30B-A3B-Instruct-2507","name":"Qwen: Qwen3 30B A3B Instruct 2507","created":1753806965,"description":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and agentic tool use. Post-trained on instruction data, it demonstrates competitive performance across reasoning (AIME, ZebraLogic), coding (MultiPL-E, LiveCodeBench), and alignment (IFEval, WritingBench) benchmarks. It outperforms its non-instruct variant on subjective and open-ended tasks while retaining strong factual and coding performance.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000009","completion":"0.0000003"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"alibaba/tongyi-deepresearch-30b-a3b","canonical_slug":"alibaba/tongyi-deepresearch-30b-a3b","hugging_face_id":"Alibaba-NLP/Tongyi-DeepResearch-30B-A3B","name":"Tongyi DeepResearch 30B A3B","created":1758210804,"description":"Tongyi DeepResearch is an agentic large language model developed by Tongyi Lab, with 30 billion total parameters activating only 3 billion per token. It's optimized for long-horizon, deep information-seeking tasks and delivers state-of-the-art performance on benchmarks like Humanity's Last Exam, BrowserComp, BrowserComp-ZH, WebWalkerQA, GAIA, xbench-DeepSearch, and FRAMES. This makes it superior for complex agentic search, reasoning, and multi-step problem-solving compared to prior models.\n\nThe model includes a fully automated synthetic data pipeline for scalable pre-training, fine-tuning, and reinforcement learning. It uses large-scale continual pre-training on diverse agentic data to boost reasoning and stay fresh. It also features end-to-end on-policy RL with a customized Group Relative Policy Optimization, including token-level gradients and negative sample filtering for stable training. The model supports ReAct for core ability checks and an IterResearch-based 'Heavy' mode for max performance through test-time scaling. It's ideal for advanced research agents, tool use, and heavy inference workflows.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000009","completion":"0.00000045","input_cache_read":"0.00000009"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen3-235b-a22b-2507","canonical_slug":"qwen/qwen3-235b-a22b-07-25","hugging_face_id":"Qwen/Qwen3-235B-A22B-Instruct-2507","name":"Qwen: Qwen3 235B A22B Instruct 2507","created":1753119555,"description":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following, logical reasoning, math, code, and tool usage. The model supports a native 262K context length and does not implement \"thinking mode\" (<think> blocks).\n\nCompared to its base variant, this version delivers significant gains in knowledge coverage, long-context reasoning, coding benchmarks, and alignment with open-ended tasks. It is particularly strong on multilingual understanding, math reasoning (e.g., AIME, HMMT), and alignment evaluations like Arena-Hard and WritingBench.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000009","completion":"0.00000058"},"top_provider":{"context_length":131072,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","reasoning_effort","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"mistralai/mistral-small-3.2-24b-instruct","canonical_slug":"mistralai/mistral-small-3.2-24b-instruct-2506","hugging_face_id":"mistralai/Mistral-Small-3.2-24B-Instruct-2506","name":"Mistral: Mistral Small 3.2 24B","created":1750443016,"description":"Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling. Compared to the 3.1 release, version 3.2 significantly improves accuracy on WildBench and Arena Hard, reduces infinite generations, and delivers gains in tool use and structured output tasks.\n\nIt supports image and text inputs with structured outputs, function/tool calling, and strong performance across coding (HumanEval+, MBPP), STEM (MMLU, MATH, GPQA), and vision benchmarks (ChartQA, DocVQA).","context_length":128000,"architecture":{"modality":"text+image->text","input_modalities":["image","text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.00000009","completion":"0.0000006"},"top_provider":{"context_length":131072,"max_completion_tokens":15000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"mistralai/ministral-3b-2512","canonical_slug":"mistralai/ministral-3b-2512","hugging_face_id":"mistralai/Ministral-3-3B-Instruct-2512","name":"Mistral: Ministral 3 3B 2512","created":1764681560,"description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","context_length":128000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000001","input_cache_read":"0.00000001"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":0.3,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"rekaai/reka-edge","canonical_slug":"rekaai/reka-edge-2603","hugging_face_id":"RekaAI/reka-edge-2603","name":"Reka Edge","created":1774026965,"description":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs. This model is optimized specifically to deliver industry-leading performance in image understanding, video analysis, object detection, and agentic tool-use.","context_length":16384,"architecture":{"modality":"text+image+video->text","input_modalities":["image","text","video"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000001"},"top_provider":{"context_length":16384,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"mistralai/devstral-small","canonical_slug":"mistralai/devstral-small-2507","hugging_face_id":"mistralai/Devstral-Small-2507","name":"Mistral: Devstral Small 1.1","created":1752160751,"description":"Devstral Small 1.1 is a 24B parameter open-weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI. Finetuned from Mistral Small 3.1 and released under the Apache 2.0 license, it features a 128k token context window and supports both Mistral-style function calling and XML output formats.\n\nDesigned for agentic coding workflows, Devstral Small 1.1 is optimized for tasks such as codebase exploration, multi-file edits, and integration into autonomous development agents like OpenHands and Cline. It achieves 53.6% on SWE-Bench Verified, surpassing all other open models on this benchmark, while remaining lightweight enough to run on a single 4090 GPU or Apple silicon machine. The model uses a Tekken tokenizer with a 131k vocabulary and is deployable via vLLM, Transformers, Ollama, LM Studio, and other OpenAI-compatible runtimes.\n","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000003","input_cache_read":"0.00000001"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"mistralai/voxtral-small-24b-2507","canonical_slug":"mistralai/voxtral-small-24b-2507","hugging_face_id":"mistralai/Voxtral-Small-24B-2507","name":"Mistral: Voxtral Small 24B 2507","created":1761835144,"description":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio is priced at $100 per million seconds.","context_length":32000,"architecture":{"modality":"text+audio->text","input_modalities":["text","audio"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000003","audio":"0.0001","input_cache_read":"0.00000001"},"top_provider":{"context_length":32000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.2,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-4.1-nano","canonical_slug":"openai/gpt-4.1-nano-2025-04-14","hugging_face_id":"","name":"OpenAI: GPT-4.1 Nano","created":1744651369,"description":"For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million token context window, and scores 80.1% on MMLU, 50.3% on GPQA, and 9.8% on Aider polyglot coding – even higher than GPT‑4o mini. It’s ideal for tasks like classification or autocompletion.","context_length":1047576,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000004","web_search":"0.01","input_cache_read":"0.000000025"},"top_provider":{"context_length":1047576,"max_completion_tokens":32768,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"bytedance-seed/seed-2.0-mini","canonical_slug":"bytedance-seed/seed-2.0-mini-20260224","hugging_face_id":"","name":"ByteDance Seed: Seed-2.0-Mini","created":1772131107,"description":"Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment. It delivers performance comparable to ByteDance-Seed-1.6, supports 256k context, four reasoning effort modes (minimal/low/medium/high), multimodal understanding, and is optimized for lightweight tasks where cost and speed take priority.","context_length":256000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000004"},"top_provider":{"context_length":262144,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-2.5-flash-lite","canonical_slug":"google/gemini-2.5-flash-lite","hugging_face_id":"","name":"Google: Gemini 2.5 Flash Lite","created":1753200276,"description":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance across common benchmarks compared to earlier Flash models. By default, \"thinking\" (i.e. multi-pass reasoning) is disabled to prioritize speed, but developers can enable it via the [Reasoning API parameter](https://openrouter.ai/docs/use-cases/reasoning-tokens) to selectively trade off cost for intelligence. ","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000004","image":"0.0000001","audio":"0.0000003","internal_reasoning":"0.0000004","input_cache_read":"0.00000001","input_cache_write":"0.00000008333333333333334"},"top_provider":{"context_length":1048576,"max_completion_tokens":65535,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"google/gemini-2.0-flash-001","canonical_slug":"google/gemini-2.0-flash-001","hugging_face_id":"","name":"Google: Gemini 2.0 Flash","created":1738769413,"description":"Gemini Flash 2.0 offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro 1.5](/google/gemini-pro-1.5). It introduces notable enhancements in multimodal understanding, coding capabilities, complex instruction following, and function calling. These advancements come together to deliver more seamless and robust agentic experiences.","context_length":1000000,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000004","image":"0.0000001","audio":"0.0000007","internal_reasoning":"0.0000004","input_cache_read":"0.000000025","input_cache_write":"0.00000008333333333333334"},"top_provider":{"context_length":1048576,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-08-31","expiration_date":"2026-06-01"},{"id":"google/gemini-2.5-flash-lite-preview-09-2025","canonical_slug":"google/gemini-2.5-flash-lite-preview-09-2025","hugging_face_id":"","name":"Google: Gemini 2.5 Flash Lite Preview 09-2025","created":1758819686,"description":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance across common benchmarks compared to earlier Flash models. By default, \"thinking\" (i.e. multi-pass reasoning) is disabled to prioritize speed, but developers can enable it via the [Reasoning API parameter](https://openrouter.ai/docs/use-cases/reasoning-tokens) to selectively trade off cost for intelligence. ","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000004","image":"0.0000001","audio":"0.0000003","internal_reasoning":"0.0000004","input_cache_read":"0.00000001","input_cache_write":"0.00000008333333333333334"},"top_provider":{"context_length":1048576,"max_completion_tokens":65535,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"qwen/qwen3-next-80b-a3b-instruct","canonical_slug":"qwen/qwen3-next-80b-a3b-instruct-2509","hugging_face_id":"Qwen/Qwen3-Next-80B-A3B-Instruct","name":"Qwen: Qwen3 Next 80B A3B Instruct","created":1757612213,"description":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks across reasoning, code generation, knowledge QA, and multilingual use, while remaining robust on alignment and formatting. Compared with prior Qwen3 instruct variants, it focuses on higher throughput and stability on ultra-long inputs and multi-turn dialogues, making it well-suited for RAG, tool use, and agentic workflows that require consistent final answers rather than visible chain-of-thought.\n\nThe model employs scaling-efficient training and decoding to improve parameter efficiency and inference speed, and has been validated on a broad set of public benchmarks where it reaches or approaches larger Qwen3 systems in several categories while outperforming earlier mid-sized baselines. It is best used as a general assistant, code helper, and long-context task solver in production settings where deterministic, instruction-following outputs are preferred.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.0000001","completion":"0.0000011"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-09-30","expiration_date":null},{"id":"qwen/qwen3-vl-8b-thinking","canonical_slug":"qwen/qwen3-vl-8b-thinking","hugging_face_id":"Qwen/Qwen3-VL-8B-Thinking","name":"Qwen: Qwen3 VL 8B Thinking","created":1760463746,"description":"Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences. It integrates enhanced multimodal alignment and long-context processing (native 256K, expandable to 1M tokens) for tasks such as scientific visual analysis, causal inference, and mathematical reasoning over image or video inputs.\n\nCompared to the Instruct edition, the Thinking version introduces deeper visual-language fusion and deliberate reasoning pathways that improve performance on long-chain logic tasks, STEM problem-solving, and multi-step video understanding. It achieves stronger temporal grounding via Interleaved-MRoPE and timestamp-aware embeddings, while maintaining robust OCR, multilingual comprehension, and text generation on par with large text-only LLMs.","context_length":256000,"architecture":{"modality":"text+image->text","input_modalities":["image","text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.000000117","completion":"0.000001365"},"top_provider":{"context_length":131072,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","presence_penalty","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":1,"top_p":0.95},"knowledge_cutoff":null,"expiration_date":null},{"id":"minimax/minimax-m2.5","canonical_slug":"minimax/minimax-m2.5-20260211","hugging_face_id":"MiniMaxAI/MiniMax-M2.5","name":"MiniMax: MiniMax M2.5","created":1770908502,"description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1 to extend into general office work, reaching fluency in generating and operating Word, Excel, and Powerpoint files, context switching between diverse software environments, and working across different agent and human teams. Scoring 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, M2.5 is also more token efficient than previous generations, having been trained to optimize its actions and output through planning.","context_length":204800,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.000000118","completion":"0.00000099","input_cache_read":"0.000000059"},"top_provider":{"context_length":196608,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","parallel_tool_calls","presence_penalty","reasoning","reasoning_effort","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"meta-llama/llama-3.3-70b-instruct","canonical_slug":"meta-llama/llama-3.3-70b-instruct","hugging_face_id":"meta-llama/Llama-3.3-70B-Instruct","name":"Meta: Llama 3.3 70B Instruct","created":1733506137,"description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model is optimized for multilingual dialogue use cases and outperforms many of the available open source and closed chat models on common industry benchmarks.\n\nSupported languages: English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai.\n\n[Model Card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md)","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"llama3"},"pricing":{"prompt":"0.00000012","completion":"0.00000038"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"qwen/qwen3-coder-next","canonical_slug":"qwen/qwen3-coder-next-2025-02-03","hugging_face_id":"Qwen/Qwen3-Coder-Next","name":"Qwen: Qwen3 Coder Next","created":1770164101,"description":"Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per token, delivering performance comparable to models with 10 to 20x higher active compute, which makes it well suited for cost-sensitive, always-on agent deployment.\n\nThe model is trained with a strong agentic focus and performs reliably on long-horizon coding tasks, complex tool usage, and recovery from execution failures. With a native 256k context window, it integrates cleanly into real-world CLI and IDE environments and adapts well to common agent scaffolds used by modern coding tools. The model operates exclusively in non-thinking mode and does not emit <think> blocks, simplifying integration for production coding agents.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen","instruct_type":null},"pricing":{"prompt":"0.00000012","completion":"0.00000075","input_cache_read":"0.00000006"},"top_provider":{"context_length":262144,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemma-4-26b-a4b-it","canonical_slug":"google/gemma-4-26b-a4b-it-20260403","hugging_face_id":"google/gemma-4-26B-A4B-it","name":"Google: Gemma 4 26B A4B ","created":1775227989,"description":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at a fraction of the compute cost. Supports multimodal input including text, images, and video (up to 60s at 1fps). Features a 256K token context window, native function calling, configurable thinking/reasoning mode, and structured output support. Released under Apache 2.0.","context_length":262144,"architecture":{"modality":"text+image+video->text","input_modalities":["image","text","video"],"output_modalities":["text"],"tokenizer":"Gemma","instruct_type":null},"pricing":{"prompt":"0.00000013","completion":"0.0000004"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"top_k":64},"knowledge_cutoff":null,"expiration_date":null},{"id":"nex-agi/deepseek-v3.1-nex-n1","canonical_slug":"nex-agi/deepseek-v3.1-nex-n1","hugging_face_id":"nex-agi/DeepSeek-V3.1-Nex-N1","name":"Nex AGI: DeepSeek V3.1 Nex N1","created":1765204393,"description":"DeepSeek V3.1 Nex-N1 is the flagship release of the Nex-N1 series — a post-trained model designed to highlight agent autonomy, tool use, and real-world productivity. \n\nNex-N1 demonstrates competitive performance across all evaluation scenarios, showing particularly strong results in practical coding and HTML generation tasks.","context_length":8192,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":null},"pricing":{"prompt":"0.000000135","completion":"0.0000005"},"top_provider":{"context_length":131072,"max_completion_tokens":163840,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","response_format","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"nousresearch/hermes-2-pro-llama-3-8b","canonical_slug":"nousresearch/hermes-2-pro-llama-3-8b","hugging_face_id":"NousResearch/Hermes-2-Pro-Llama-3-8B","name":"NousResearch: Hermes 2 Pro - Llama-3 8B","created":1716768000,"description":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mode dataset developed in-house.","context_length":8192,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"chatml"},"pricing":{"prompt":"0.00000014","completion":"0.00000014"},"top_provider":{"context_length":8192,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"google/gemma-4-31b-it","canonical_slug":"google/gemma-4-31b-it-20260402","hugging_face_id":"google/gemma-4-31B-it","name":"Google: Gemma 4 31B","created":1775148486,"description":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function calling, and multilingual support across 140+ languages. Strong on coding, reasoning, and document understanding tasks. Apache 2.0 license.","context_length":262144,"architecture":{"modality":"text+image+video->text","input_modalities":["image","text","video"],"output_modalities":["text"],"tokenizer":"Gemma","instruct_type":null},"pricing":{"prompt":"0.00000014","completion":"0.0000004"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"top_k":64,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"qwen/qwen3-30b-a3b","canonical_slug":"qwen/qwen3-30b-a3b-04-28","hugging_face_id":"Qwen/Qwen3-30B-A3B","name":"Qwen: Qwen3 30B A3B","created":1745878604,"description":"Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks. Its unique ability to switch seamlessly between a thinking mode for complex reasoning and a non-thinking mode for efficient dialogue ensures versatile, high-quality performance.\n\nSignificantly outperforming prior models like QwQ and Qwen2.5, Qwen3 delivers superior mathematics, coding, commonsense reasoning, creative writing, and interactive dialogue capabilities. The Qwen3-30B-A3B variant includes 30.5 billion parameters (3.3 billion activated), 48 layers, 128 experts (8 activated per task), and supports up to 131K token contexts with YaRN, setting a new standard among open-source models.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":"qwen3"},"pricing":{"prompt":"0.00000014","completion":"0.00000055"},"top_provider":{"context_length":32768,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"tencent/hunyuan-a13b-instruct","canonical_slug":"tencent/hunyuan-a13b-instruct","hugging_face_id":"tencent/Hunyuan-A13B-Instruct","name":"Tencent: Hunyuan A13B Instruct","created":1751987664,"description":"Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought. It offers competitive benchmark performance across mathematics, science, coding, and multi-turn reasoning tasks, while maintaining high inference efficiency via Grouped Query Attention (GQA) and quantization support (FP8, GPTQ, etc.).","context_length":32768,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000014","completion":"0.00000057"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","reasoning","response_format","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"essentialai/rnj-1-instruct","canonical_slug":"essentialai/rnj-1-instruct","hugging_face_id":"EssentialAI/rnj-1-instruct","name":"EssentialAI: Rnj 1 Instruct","created":1765094847,"description":"Rnj-1 is an 8B-parameter, dense, open-weight model family developed by Essential AI and trained from scratch with a focus on programming, math, and scientific reasoning. The model demonstrates strong performance across multiple programming languages, tool-use workflows, and agentic execution environments (e.g., mini-SWE-agent). ","context_length":32768,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.00000015"},"top_provider":{"context_length":32768,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"mistralai/ministral-8b-2512","canonical_slug":"mistralai/ministral-8b-2512","hugging_face_id":"mistralai/Ministral-3-8B-Instruct-2512","name":"Mistral: Ministral 3 8B 2512","created":1764681654,"description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","context_length":128000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.00000015","input_cache_read":"0.000000015"},"top_provider":{"context_length":262144,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":0.3,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"allenai/olmo-3-32b-think","canonical_slug":"allenai/olmo-3-32b-think-20251121","hugging_face_id":"allenai/Olmo-3-32B-Think","name":"AllenAI: Olmo 3 32B Think","created":1763758276,"description":"Olmo 3 32B Think is a large-scale, 32-billion-parameter model purpose-built for deep reasoning, complex logic chains and advanced instruction-following scenarios. Its capacity enables strong performance on demanding evaluation tasks and highly nuanced conversational reasoning. Developed by Ai2 under the Apache 2.0 license, Olmo 3 32B Think embodies the Olmo initiative’s commitment to openness, offering full transparency across weights, code and training methodology.","context_length":65536,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000005"},"top_provider":{"context_length":65536,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"allenai/olmo-3.1-32b-think","canonical_slug":"allenai/olmo-3.1-32b-think-20251215","hugging_face_id":"allenai/Olmo-3.1-32B-Think","name":"AllenAI: Olmo 3.1 32B Think","created":1765907719,"description":"Olmo 3.1 32B Think is a large-scale, 32-billion-parameter model designed for deep reasoning, complex multi-step logic, and advanced instruction following. Building on the Olmo 3 series, version 3.1 delivers refined reasoning behavior and stronger performance across demanding evaluations and nuanced conversational tasks. Developed by Ai2 under the Apache 2.0 license, Olmo 3.1 32B Think continues the Olmo initiative’s commitment to openness, providing full transparency across model weights, code, and training methodology.","context_length":65536,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000005"},"top_provider":{"context_length":65536,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":"2026-04-06"},{"id":"meta-llama/llama-4-maverick","canonical_slug":"meta-llama/llama-4-maverick-17b-128e-instruct","hugging_face_id":"meta-llama/Llama-4-Maverick-17B-128E-Instruct","name":"Meta: Llama 4 Maverick","created":1743881822,"description":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward pass (400B total). It supports multilingual text and image input, and produces multilingual text and code output across 12 supported languages. Optimized for vision-language tasks, Maverick is instruction-tuned for assistant-like behavior, image reasoning, and general-purpose multimodal interaction.\n\nMaverick features early fusion for native multimodality and a 1 million token context window. It was trained on a curated mixture of public, licensed, and Meta-platform data, covering ~22 trillion tokens, with a knowledge cutoff in August 2024. Released on April 5, 2025 under the Llama 4 Community License, Maverick is suited for research and commercial applications requiring advanced multimodal understanding and high model throughput.","context_length":1048576,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Llama4","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006"},"top_provider":{"context_length":1048576,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-08-31","expiration_date":null},{"id":"openai/gpt-4o-mini-search-preview","canonical_slug":"openai/gpt-4o-mini-search-preview-2025-03-11","hugging_face_id":"","name":"OpenAI: GPT-4o-mini Search Preview","created":1741818122,"description":"GPT-4o mini Search Preview is a specialized model for web search in Chat Completions. It is trained to understand and execute web search queries.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006","web_search":"0.0275"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","structured_outputs","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"mistralai/mistral-small-2603","canonical_slug":"mistralai/mistral-small-2603","hugging_face_id":"mistralai/Mistral-Small-4-119B-2603","name":"Mistral: Mistral Small 4","created":1773695685,"description":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system. It combines strong reasoning from Magistral, multimodal understanding from Pixtral, and agentic coding capabilities from Devstral, enabling one model to handle complex analysis, software development, and visual tasks within the same workflow.","context_length":262144,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006","input_cache_read":"0.000000015"},"top_provider":{"context_length":262144,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"upstage/solar-pro-3","canonical_slug":"upstage/solar-pro-3","hugging_face_id":"","name":"Upstage: Solar Pro 3","created":1769481200,"description":"Solar Pro 3 is Upstage's powerful Mixture-of-Experts (MoE) language model. With 102B total parameters and 12B active parameters per forward pass, it delivers exceptional performance while maintaining computational efficiency. Optimized for Korean with English and Japanese support.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006","input_cache_read":"0.000000015"},"top_provider":{"context_length":128000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","structured_outputs","temperature","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-4o-mini-2024-07-18","canonical_slug":"openai/gpt-4o-mini-2024-07-18","hugging_face_id":null,"name":"OpenAI: GPT-4o-mini (2024-07-18)","created":1721260800,"description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.\n\nAs their most advanced small model, it is many multiples more affordable than other recent frontier models, and more than 60% cheaper than [GPT-3.5 Turbo](/models/openai/gpt-3.5-turbo). It maintains SOTA intelligence, while being significantly more cost-effective.\n\nGPT-4o mini achieves an 82% score on MMLU and presently ranks higher than GPT-4 on chat preferences [common leaderboards](https://arena.lmsys.org/).\n\nCheck out the [launch announcement](https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/) to learn more.\n\n#multimodal","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006","input_cache_read":"0.000000075"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/gpt-4o-mini","canonical_slug":"openai/gpt-4o-mini","hugging_face_id":null,"name":"OpenAI: GPT-4o-mini","created":1721260800,"description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.\n\nAs their most advanced small model, it is many multiples more affordable than other recent frontier models, and more than 60% cheaper than [GPT-3.5 Turbo](/models/openai/gpt-3.5-turbo). It maintains SOTA intelligence, while being significantly more cost-effective.\n\nGPT-4o mini achieves an 82% score on MMLU and presently ranks higher than GPT-4 on chat preferences [common leaderboards](https://arena.lmsys.org/).\n\nCheck out the [launch announcement](https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/) to learn more.\n\n#multimodal","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006","input_cache_read":"0.000000075"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"cohere/command-r-08-2024","canonical_slug":"cohere/command-r-08-2024","hugging_face_id":null,"name":"Cohere: Command R (08-2024)","created":1724976000,"description":"command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use. More broadly, it is better at math, code and reasoning and is competitive with the previous version of the larger Command R+ model.\n\nRead the launch post [here](https://docs.cohere.com/changelog/command-gets-refreshed).\n\nUse of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Cohere","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006"},"top_provider":{"context_length":128000,"max_completion_tokens":4000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-03-31","expiration_date":null},{"id":"qwen/qwen3-vl-30b-a3b-instruct","canonical_slug":"qwen/qwen3-vl-30b-a3b-instruct","hugging_face_id":"Qwen/Qwen3-VL-30B-A3B-Instruct","name":"Qwen: Qwen3 VL 30B A3B Instruct","created":1759794476,"description":"Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Instruct variant optimizes instruction-following for general multimodal tasks. It excels in perception of real-world/synthetic categories, 2D/3D spatial grounding, and long-form visual comprehension, achieving competitive multimodal benchmark results. For agentic use, it handles multi-image multi-turn instructions, video timeline alignments, GUI automation, and visual coding from sketches to debugged UI. Text performance matches flagship Qwen3 models, suiting document AI, OCR, UI assistance, spatial tasks, and agent research.","context_length":262144,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000006","input_cache_read":"0.00000015"},"top_provider":{"context_length":128000,"max_completion_tokens":32000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.7,"top_p":0.8,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":1},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen3-next-80b-a3b-thinking","canonical_slug":"qwen/qwen3-next-80b-a3b-thinking-2509","hugging_face_id":"Qwen/Qwen3-Next-80B-A3B-Thinking","name":"Qwen: Qwen3 Next 80B A3B Thinking","created":1757612284,"description":"Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default. It’s designed for hard multi-step problems; math proofs, code synthesis/debugging, logic, and agentic planning, and reports strong results across knowledge, reasoning, coding, alignment, and multilingual evaluations. Compared with prior Qwen3 variants, it emphasizes stability under long chains of thought and efficient scaling during inference, and it is tuned to follow complex instructions while reducing repetitive or off-task behavior.\n\nThe model is suitable for agent frameworks and tool use (function calling), retrieval-heavy workflows, and standardized benchmarking where step-by-step solutions are required. It supports long, detailed completions and leverages throughput-oriented techniques (e.g., multi-token prediction) for faster generation. Note that it operates in thinking-only mode.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000015","completion":"0.0000012"},"top_provider":{"context_length":128000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-09-30","expiration_date":null},{"id":"qwen/qwen3.5-35b-a3b","canonical_slug":"qwen/qwen3.5-35b-a3b-20260224","hugging_face_id":"Qwen/Qwen3.5-35B-A3B","name":"Qwen: Qwen3.5-35B-A3B","created":1772053822,"description":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall performance is comparable to that of the Qwen3.5-27B.","context_length":256000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.0000001625","completion":"0.0000013"},"top_provider":{"context_length":262144,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"thedrummer/rocinante-12b","canonical_slug":"thedrummer/rocinante-12b","hugging_face_id":"TheDrummer/Rocinante-12B-v1.1","name":"TheDrummer: Rocinante 12B","created":1727654400,"description":"Rocinante 12B is designed for engaging storytelling and rich prose.\n\nEarly testers have reported:\n- Expanded vocabulary with unique and expressive word choices\n- Enhanced creativity for vivid narratives\n- Adventure-filled and captivating stories","context_length":32768,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen","instruct_type":"chatml"},"pricing":{"prompt":"0.00000017","completion":"0.00000043"},"top_provider":{"context_length":32768,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-04-30","expiration_date":null},{"id":"qwen/qwen3.5-27b","canonical_slug":"qwen/qwen3.5-27b-20260224","hugging_face_id":"Qwen/Qwen3.5-27B","name":"Qwen: Qwen3.5-27B","created":1772053810,"description":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance. Its overall capabilities are comparable to those of the Qwen3.5-122B-A10B.","context_length":256000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.000000195","completion":"0.00000156"},"top_provider":{"context_length":262144,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"mistralai/ministral-14b-2512","canonical_slug":"mistralai/ministral-14b-2512","hugging_face_id":"mistralai/Ministral-3-14B-Instruct-2512","name":"Mistral: Ministral 3 14B 2512","created":1764681735,"description":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart. A powerful and efficient language model with vision capabilities.","context_length":128000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000002","input_cache_read":"0.00000002"},"top_provider":{"context_length":262144,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":0.3,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"x-ai/grok-4.1-fast","canonical_slug":"x-ai/grok-4.1-fast","hugging_face_id":"","name":"xAI: Grok 4.1 Fast","created":1763587502,"description":"Grok 4.1 Fast is xAI's best agentic tool calling model that shines in real-world use cases like customer support and deep research. 2M context window.\n\nReasoning can be enabled/disabled using the `reasoning` `enabled` parameter in the API. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#controlling-reasoning-tokens)","context_length":2000000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000005","web_search":"0.005","input_cache_read":"0.00000005"},"top_provider":{"context_length":2000000,"max_completion_tokens":30000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","logprobs","max_tokens","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":0.7,"top_p":0.95,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"x-ai/grok-4-fast","canonical_slug":"x-ai/grok-4-fast","hugging_face_id":"","name":"xAI: Grok 4 Fast","created":1758240090,"description":"Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency and a 2M token context window. It comes in two flavors: non-reasoning and reasoning. Read more about the model on xAI's [news post](http://x.ai/news/grok-4-fast).\n\nReasoning can be enabled/disabled using the `reasoning` `enabled` parameter in the API. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#controlling-reasoning-tokens)","context_length":2000000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000005","web_search":"0.005","input_cache_read":"0.00000005"},"top_provider":{"context_length":2000000,"max_completion_tokens":30000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","logprobs","max_tokens","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-09-30","expiration_date":null},{"id":"mistralai/mistral-saba","canonical_slug":"mistralai/mistral-saba-2502","hugging_face_id":"","name":"Mistral: Saba","created":1739803239,"description":"Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. Trained on curated regional datasets, it supports multiple Indian-origin languages—including Tamil and Malayalam—alongside Arabic. This makes it a versatile option for a range of regional and multilingual applications. Read more at the blog post [here](https://mistral.ai/en/news/mistral-saba)","context_length":32000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000006","input_cache_read":"0.00000002"},"top_provider":{"context_length":32768,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2024-09-30","expiration_date":null},{"id":"allenai/olmo-3.1-32b-instruct","canonical_slug":"allenai/olmo-3.1-32b-instruct-20251215","hugging_face_id":"allenai/Olmo-3.1-32B-Instruct","name":"AllenAI: Olmo 3.1 32B Instruct","created":1767728554,"description":"Olmo 3.1 32B Instruct is a large-scale, 32-billion-parameter instruction-tuned language model engineered for high-performance conversational AI, multi-turn dialogue, and practical instruction following. As part of the Olmo 3.1 family, this variant emphasizes responsiveness to complex user directions and robust chat interactions while retaining strong capabilities on reasoning and coding benchmarks. Developed by Ai2 under the Apache 2.0 license, Olmo 3.1 32B Instruct reflects the Olmo initiative’s commitment to openness and transparency.","context_length":65536,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000006"},"top_provider":{"context_length":65536,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"meituan/longcat-flash-chat","canonical_slug":"meituan/longcat-flash-chat","hugging_face_id":"meituan-longcat/LongCat-Flash-Chat","name":"Meituan: LongCat Flash Chat","created":1757427658,"description":"LongCat-Flash-Chat is a large-scale Mixture-of-Experts (MoE) model with 560B total parameters, of which 18.6B–31.3B (≈27B on average) are dynamically activated per input. It introduces a shortcut-connected MoE design to reduce communication overhead and achieve high throughput while maintaining training stability through advanced scaling strategies such as hyperparameter transfer, deterministic computation, and multi-stage optimization.\n\nThis release, LongCat-Flash-Chat, is a non-thinking foundation model optimized for conversational and agentic tasks. It supports long context windows up to 128K tokens and shows competitive performance across reasoning, coding, instruction following, and domain benchmarks, with particular strengths in tool use and complex multi-step interactions.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000008","input_cache_read":"0.0000002"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen3-vl-30b-a3b-thinking","canonical_slug":"qwen/qwen3-vl-30b-a3b-thinking","hugging_face_id":"Qwen/Qwen3-VL-30B-A3B-Thinking","name":"Qwen: Qwen3 VL 30B A3B Thinking","created":1759794479,"description":"Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Thinking variant enhances reasoning in STEM, math, and complex tasks. It excels in perception of real-world/synthetic categories, 2D/3D spatial grounding, and long-form visual comprehension, achieving competitive multimodal benchmark results. For agentic use, it handles multi-image multi-turn instructions, video timeline alignments, GUI automation, and visual coding from sketches to debugged UI. Text performance matches flagship Qwen3 models, suiting document AI, OCR, UI assistance, spatial tasks, and agent research.","context_length":0,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.000001"},"top_provider":{"context_length":131072,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.8,"top_p":0.95,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":1},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"prime-intellect/intellect-3","canonical_slug":"prime-intellect/intellect-3-20251126","hugging_face_id":"PrimeIntellect/INTELLECT-3-FP8","name":"Prime Intellect: INTELLECT-3","created":1764212534,"description":"INTELLECT-3 is a 106B-parameter Mixture-of-Experts model (12B active) post-trained from GLM-4.5-Air-Base using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL). It offers state-of-the-art performance for its size across math, code, science, and general reasoning, consistently outperforming many larger frontier models. Designed for strong multi-step problem solving, it maintains high accuracy on structured tasks while remaining efficient at inference thanks to its MoE architecture.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000011"},"top_provider":{"context_length":131072,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"z-ai/glm-4.5-air","canonical_slug":"z-ai/glm-4.5-air","hugging_face_id":"zai-org/GLM-4.5-Air","name":"Z.ai: GLM 4.5 Air","created":1753471258,"description":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications. Like GLM-4.5, it adopts the Mixture-of-Experts (MoE) architecture but with a more compact parameter size. GLM-4.5-Air also supports hybrid inference modes, offering a \"thinking mode\" for advanced reasoning and tool use, and a \"non-thinking mode\" for real-time interaction. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000012"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.75,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-12-31","expiration_date":null},{"id":"openai/gpt-5.4-nano","canonical_slug":"openai/gpt-5.4-nano-20260317","hugging_face_id":"","name":"OpenAI: GPT-5.4 Nano","created":1773748187,"description":"GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks. It supports text and image inputs and is designed for low-latency use cases such as classification, data extraction, ranking, and sub-agent execution.\n\nThe model prioritizes responsiveness and efficiency over deep reasoning, making it ideal for pipelines that require fast, reliable outputs at scale. GPT-5.4 nano is well suited for background tasks, real-time systems, and distributed agent architectures where minimizing cost and latency is essential.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.00000125","web_search":"0.01","input_cache_read":"0.00000002"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-08-31","expiration_date":null},{"id":"x-ai/grok-code-fast-1","canonical_slug":"x-ai/grok-code-fast-1","hugging_face_id":"","name":"xAI: Grok Code Fast 1","created":1756238927,"description":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding. With reasoning traces visible in the response, developers can steer Grok Code for high-quality work flows.","context_length":256000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.0000002","completion":"0.0000015","web_search":"0.005","input_cache_read":"0.00000002"},"top_provider":{"context_length":256000,"max_completion_tokens":10000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","logprobs","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-09-30","expiration_date":null},{"id":"qwen/qwen3-vl-235b-a22b-instruct","canonical_slug":"qwen/qwen3-vl-235b-a22b-instruct","hugging_face_id":"Qwen/Qwen3-VL-235B-A22B-Instruct","name":"Qwen: Qwen3 VL 235B A22B Instruct","created":1758668687,"description":"Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video. The Instruct model targets general vision-language use (VQA, document parsing, chart/table extraction, multilingual OCR). The series emphasizes robust perception (recognition of diverse real-world and synthetic categories), spatial understanding (2D/3D grounding), and long-form visual comprehension, with competitive results on public multimodal benchmarks for both perception and reasoning.\n\nBeyond analysis, Qwen3-VL supports agentic interaction and tool use: it can follow complex instructions over multi-image, multi-turn dialogues; align text to video timelines for precise temporal queries; and operate GUI elements for automation tasks. The models also enable visual coding workflows—turning sketches or mockups into code and assisting with UI debugging—while maintaining strong text-only performance comparable to the flagship Qwen3 language models. This makes Qwen3-VL suitable for production scenarios spanning document AI, multilingual OCR, software/UI assistance, spatial/embodied tasks, and research on vision-language agents.","context_length":131072,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000021","completion":"0.0000019"},"top_provider":{"context_length":131072,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.7,"top_p":0.8,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"deepseek/deepseek-chat-v3-0324","canonical_slug":"deepseek/deepseek-chat-v3-0324","hugging_face_id":"deepseek-ai/DeepSeek-V3-0324","name":"DeepSeek: DeepSeek V3 0324","created":1742824755,"description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.\n\nIt succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) model and performs really well on a variety of tasks.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":null},"pricing":{"prompt":"0.000000216","completion":"0.00000088","input_cache_read":"0.00000018"},"top_provider":{"context_length":131072,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-07-31","expiration_date":null},{"id":"arcee-ai/trinity-large-thinking","canonical_slug":"arcee-ai/trinity-large-thinking","hugging_face_id":"arcee-ai/Trinity-Large-Thinking","name":"Arcee AI: Trinity Large Thinking","created":1775058318,"description":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI. It shows strong performance in PinchBench, agentic workloads, and reasoning tasks. It is free in open claw for the first five days. Launch video: https://youtu.be/Gc82AXLa0Rg?si=4RLn6WBz33qT--B7","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000022","completion":"0.00000085"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.3,"top_p":0.8,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"qwen/qwen3-coder","canonical_slug":"qwen/qwen3-coder-480b-a35b-07-25","hugging_face_id":"Qwen/Qwen3-Coder-480B-A35B-Instruct","name":"Qwen: Qwen3 Coder 480B A35B","created":1753230546,"description":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over repositories. The model features 480 billion total parameters, with 35 billion active per forward pass (8 out of 160 experts).\n\nPricing for the Alibaba endpoints varies by context length. Once a request is greater than 128k input tokens, the higher pricing is used.","context_length":1048576,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000022","completion":"0.0000018"},"top_provider":{"context_length":262144,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"meta-llama/llama-4-scout","canonical_slug":"meta-llama/llama-4-scout-17b-16e-instruct","hugging_face_id":"meta-llama/Llama-4-Scout-17B-16E-Instruct","name":"Meta: Llama 4 Scout","created":1743881519,"description":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, Scout uses 16 experts per forward pass and features a context length of 10 million tokens, with a training corpus of ~40 trillion tokens.\n\nBuilt for high efficiency and local or commercial deployment, Llama 4 Scout incorporates early fusion for seamless modality integration. It is instruction-tuned for use in multilingual chat, captioning, and image understanding tasks. Released under the Llama 4 Community License, it was last trained on data up to August 2024 and launched publicly on April 5, 2025.","context_length":10000000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Llama4","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.0000007"},"top_provider":{"context_length":1310720,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-08-31","expiration_date":null},{"id":"inception/mercury","canonical_slug":"inception/mercury","hugging_face_id":"","name":"Inception: Mercury","created":1750973026,"description":"Mercury is the first diffusion large language model (dLLM). Applying a breakthrough discrete diffusion approach, the model runs 5-10x faster than even speed optimized models like GPT-4.1 Nano and Claude 3.5 Haiku while matching their performance. Mercury's speed enables developers to provide responsive user experiences, including with voice agents, search interfaces, and chatbots. Read more in the [blog post]\n(https://www.inceptionlabs.ai/blog/introducing-mercury) here. ","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.00000075","input_cache_read":"0.000000025"},"top_provider":{"context_length":128000,"max_completion_tokens":32000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","stop","structured_outputs","temperature","tool_choice","tools"],"default_parameters":{"temperature":0,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":"2026-04-15"},{"id":"inception/mercury-coder","canonical_slug":"inception/mercury-coder-small-beta","hugging_face_id":"","name":"Inception: Mercury Coder","created":1746033880,"description":"Mercury Coder is the first diffusion large language model (dLLM). Applying a breakthrough discrete diffusion approach, the model runs 5-10x faster than even speed optimized models like Claude 3.5 Haiku and GPT-4o Mini while matching their performance. Mercury Coder's speed means that developers can stay in the flow while coding, enjoying rapid chat-based iteration and responsive code completion suggestions. On Copilot Arena, Mercury Coder ranks 1st in speed and ties for 2nd in quality. Read more in the [blog post here](https://www.inceptionlabs.ai/blog/introducing-mercury).","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.00000075","input_cache_read":"0.000000025"},"top_provider":{"context_length":128000,"max_completion_tokens":32000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","stop","structured_outputs","temperature","tool_choice","tools"],"default_parameters":{"temperature":0,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":"2026-04-15"},{"id":"inception/mercury-2","canonical_slug":"inception/mercury-2-20260304","hugging_face_id":null,"name":"Inception: Mercury 2","created":1772636275,"description":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM).\nInstead of generating tokens sequentially, Mercury 2 produces and refines multiple tokens in parallel, achieving >1,000 tokens/sec on standard GPUs. Mercury 2 is 5x+ faster than leading speed-optimized LLMs like Claude 4.5 Haiku and GPT 5 Mini, at a fraction of the cost. \nMercury 2 supports tunable reasoning levels, 128K context, native tool use, and schema-aligned JSON output. Built for coding workflows where latency compounds, real-time voice/search, and agent loops. OpenAI API compatible. Read more in the [blog post](https://www.inceptionlabs.ai/blog/introducing-mercury-2).","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.00000075","input_cache_read":"0.000000025"},"top_provider":{"context_length":128000,"max_completion_tokens":50000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools"],"default_parameters":{"temperature":0.75,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-3.1-flash-lite-preview","canonical_slug":"google/gemini-3.1-flash-lite-preview-20260303","hugging_face_id":"","name":"Google: Gemini 3.1 Flash Lite Preview","created":1772512673,"description":"Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases. It outperforms Gemini 2.5 Flash Lite on overall quality and approaches Gemini 2.5 Flash performance across key capabilities. Improvements span audio input/ASR, RAG snippet ranking, translation, data extraction, and code completion. Supports full thinking levels (minimal, low, medium, high) for fine-grained cost/performance trade-offs. Priced at half the cost of Gemini 3 Flash.","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","video","file","audio"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.0000015","image":"0.00000025","audio":"0.0000005","internal_reasoning":"0.0000015","input_cache_read":"0.000000025","input_cache_write":"0.00000008333333333333334"},"top_provider":{"context_length":1048576,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5-mini","canonical_slug":"openai/gpt-5-mini-2025-08-07","hugging_face_id":"","name":"OpenAI: GPT-5 Mini","created":1754587407,"description":"GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost. GPT-5 Mini is the successor to OpenAI's o4-mini model.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.000002","web_search":"0.01","input_cache_read":"0.000000025"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2024-05-31","expiration_date":null},{"id":"openai/gpt-5.1-codex-mini","canonical_slug":"openai/gpt-5.1-codex-mini-20251113","hugging_face_id":"","name":"OpenAI: GPT-5.1-Codex-Mini","created":1763057820,"description":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","context_length":400000,"architecture":{"modality":"text+image->text","input_modalities":["image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.000002","input_cache_read":"0.000000025"},"top_provider":{"context_length":400000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"bytedance-seed/seed-1.6","canonical_slug":"bytedance-seed/seed-1.6-20250625","hugging_face_id":"","name":"ByteDance Seed: Seed 1.6","created":1766504997,"description":"Seed 1.6 is a general-purpose model released by the ByteDance Seed team. It incorporates multimodal capabilities and adaptive deep thinking with a 256K context window.","context_length":262144,"architecture":{"modality":"text+image+video->text","input_modalities":["image","text","video"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.000002"},"top_provider":{"context_length":262144,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"bytedance-seed/seed-2.0-lite","canonical_slug":"bytedance-seed/seed-2.0-lite-20260309","hugging_face_id":null,"name":"ByteDance Seed: Seed-2.0-Lite","created":1773157231,"description":"Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably lower latency, making it a practical default choice for most production workloads across text, vision, and tools. Engineered for high-frequency visual understanding and agentic workflows, it's an ideal choice for deployment at scale with minimal latency.","context_length":256000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000025","completion":"0.000002"},"top_provider":{"context_length":262144,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"minimax/minimax-m2","canonical_slug":"minimax/minimax-m2","hugging_face_id":"MiniMaxAI/MiniMax-M2","name":"MiniMax: MiniMax M2","created":1761252093,"description":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion total), it delivers near-frontier intelligence across general reasoning, tool use, and multi-step task execution while maintaining low latency and deployment efficiency.\n\nThe model excels in code generation, multi-file editing, compile-run-fix loops, and test-validated repair, showing strong results on SWE-Bench Verified, Multi-SWE-Bench, and Terminal-Bench. It also performs competitively in agentic evaluations such as BrowseComp and GAIA, effectively handling long-horizon planning, retrieval, and recovery from execution errors.\n\nBenchmarked by [Artificial Analysis](https://artificialanalysis.ai/models/minimax-m2), MiniMax-M2 ranks among the top open-source models for composite intelligence, spanning mathematics, science, and instruction-following. Its small activation footprint enables fast inference, high concurrency, and improved unit economics, making it well-suited for large-scale agents, developer assistants, and reasoning-driven applications that require responsiveness and cost efficiency.\n\nTo avoid degrading this model's performance, MiniMax highly recommends preserving reasoning between turns. Learn more about using reasoning_details to pass back reasoning in our [docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#preserving-reasoning-blocks).","context_length":204800,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.000000255","completion":"0.000001","input_cache_read":"0.00000003"},"top_provider":{"context_length":196608,"max_completion_tokens":196608,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"deepseek/deepseek-v3.2","canonical_slug":"deepseek/deepseek-v3.2-20251201","hugging_face_id":"deepseek-ai/DeepSeek-V3.2","name":"DeepSeek: DeepSeek V3.2","created":1764594642,"description":"DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism that reduces training and inference cost while preserving quality in long-context scenarios. A scalable reinforcement learning post-training framework further improves reasoning, with reported performance in the GPT-5 class, and the model has demonstrated gold-medal results on the 2025 IMO and IOI. V3.2 also uses a large-scale agentic task synthesis pipeline to better integrate reasoning into tool-use settings, boosting compliance and generalization in interactive environments.\n\nUsers can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":null},"pricing":{"prompt":"0.00000026","completion":"0.00000038","input_cache_read":"0.00000013"},"top_provider":{"context_length":163840,"max_completion_tokens":163840,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"qwen/qwen-plus-2025-07-28:thinking","canonical_slug":"qwen/qwen-plus-2025-07-28","hugging_face_id":"","name":"Qwen: Qwen Plus 0728 (thinking)","created":1757347599,"description":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","context_length":1000000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000026","completion":"0.00000078"},"top_provider":{"context_length":1000000,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","presence_penalty","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen-plus-2025-07-28","canonical_slug":"qwen/qwen-plus-2025-07-28","hugging_face_id":"","name":"Qwen: Qwen Plus 0728","created":1757347599,"description":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","context_length":1000000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000026","completion":"0.00000078"},"top_provider":{"context_length":1000000,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","presence_penalty","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen3.5-plus-02-15","canonical_slug":"qwen/qwen3.5-plus-20260216","hugging_face_id":"","name":"Qwen: Qwen3.5 Plus 2026-02-15","created":1771229416,"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_length":1000000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000026","completion":"0.00000156"},"top_provider":{"context_length":1000000,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","presence_penalty","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"qwen/qwen3.5-122b-a10b","canonical_slug":"qwen/qwen3.5-122b-a10b-20260224","hugging_face_id":"Qwen/Qwen3.5-122B-A10B","name":"Qwen: Qwen3.5-122B-A10B","created":1772053789,"description":"The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of overall performance, this model is second only to Qwen3.5-397B-A17B. Its text capabilities significantly outperform those of Qwen3-235B-2507, and its visual capabilities surpass those of Qwen3-VL-235B.","context_length":262144,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000026","completion":"0.00000208"},"top_provider":{"context_length":262144,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"deepseek/deepseek-v3.2-exp","canonical_slug":"deepseek/deepseek-v3.2-exp","hugging_face_id":"deepseek-ai/DeepSeek-V3.2-Exp","name":"DeepSeek: DeepSeek V3.2 Exp","created":1759150481,"description":"DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism designed to improve training and inference efficiency in long-context scenarios while maintaining output quality. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)\n\nThe model was trained under conditions aligned with V3.1-Terminus to enable direct comparison. Benchmarking shows performance roughly on par with V3.1 across reasoning, coding, and agentic tool-use tasks, with minor tradeoffs and gains depending on the domain. This release focuses on validating architectural optimizations for extended context lengths rather than advancing raw task accuracy, making it primarily a research-oriented model for exploring efficient transformer designs.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":"deepseek-v3.1"},"pricing":{"prompt":"0.00000027","completion":"0.00000041"},"top_provider":{"context_length":163840,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":"2025-07-31","expiration_date":null},{"id":"deepseek/deepseek-chat-v3.1","canonical_slug":"deepseek/deepseek-chat-v3.1","hugging_face_id":"deepseek-ai/DeepSeek-V3.1","name":"DeepSeek: DeepSeek V3.1","created":1755779628,"description":"DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context training process, reaching up to 128K tokens, and uses FP8 microscaling for efficient inference. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)\n\nThe model improves tool use, code generation, and reasoning efficiency, achieving performance comparable to DeepSeek-R1 on difficult benchmarks while responding more quickly. It supports structured tool calling, code agents, and search agents, making it suitable for research, coding, and agentic workflows. \n\nIt succeeds the [DeepSeek V3-0324](/deepseek/deepseek-chat-v3-0324) model and performs well on a variety of tasks.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":"deepseek-v3.1"},"pricing":{"prompt":"0.00000027","completion":"0.000001","input_cache_read":"0.000000135"},"top_provider":{"context_length":163840,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"deepseek/deepseek-v3.1-terminus","canonical_slug":"deepseek/deepseek-v3.1-terminus","hugging_face_id":"deepseek-ai/DeepSeek-V3.1-Terminus","name":"DeepSeek: DeepSeek V3.1 Terminus","created":1758548275,"description":"DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's performance in coding and search agents. It is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes. It extends the DeepSeek-V3 base with a two-phase long-context training process, reaching up to 128K tokens, and uses FP8 microscaling for efficient inference. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)\n\nThe model improves tool use, code generation, and reasoning efficiency, achieving performance comparable to DeepSeek-R1 on difficult benchmarks while responding more quickly. It supports structured tool calling, code agents, and search agents, making it suitable for research, coding, and agentic workflows. ","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":"deepseek-v3.1"},"pricing":{"prompt":"0.00000027","completion":"0.000001","input_cache_read":"0.000000135"},"top_provider":{"context_length":131072,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"baidu/ernie-4.5-300b-a47b","canonical_slug":"baidu/ernie-4.5-300b-a47b","hugging_face_id":"baidu/ERNIE-4.5-300B-A47B-PT","name":"Baidu: ERNIE 4.5 300B A47B ","created":1751300139,"description":"ERNIE-4.5-300B-A47B is a 300B parameter Mixture-of-Experts (MoE) language model developed by Baidu as part of the ERNIE 4.5 series. It activates 47B parameters per token and supports text generation in both English and Chinese. Optimized for high-throughput inference and efficient scaling, it uses a heterogeneous MoE structure with advanced routing and quantization strategies, including FP8 and 2-bit formats. This version is fine-tuned for language-only tasks and supports reasoning, tool parameters, and extended context lengths up to 131k tokens. Suitable for general-purpose LLM applications with high reasoning and throughput demands.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000028","completion":"0.0000011"},"top_provider":{"context_length":123000,"max_completion_tokens":12000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen3-235b-a22b-thinking-2507","canonical_slug":"qwen/qwen3-235b-a22b-thinking-2507","hugging_face_id":"Qwen/Qwen3-235B-A22B-Thinking-2507","name":"Qwen: Qwen3 235B A22B Thinking 2507","created":1753449557,"description":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144 tokens of context. This \"thinking-only\" variant enhances structured logical reasoning, mathematics, science, and long-form generation, showing strong benchmark performance across AIME, SuperGPQA, LiveCodeBench, and MMLU-Redux. It enforces a special reasoning mode (</think>) and is designed for high-token outputs (up to 81,920 tokens) in challenging domains.\n\nThe model is instruction-tuned and excels at step-by-step reasoning, tool use, agentic workflows, and multilingual tasks. This release represents the most capable open-source variant in the Qwen3-235B series, surpassing many closed models in structured reasoning use cases.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":"qwen3"},"pricing":{"prompt":"0.00000028","completion":"0.0000023","input_cache_read":"0.00000028"},"top_provider":{"context_length":128000,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"deepseek/deepseek-r1-distill-qwen-32b","canonical_slug":"deepseek/deepseek-r1-distill-qwen-32b","hugging_face_id":"deepseek-ai/DeepSeek-R1-Distill-Qwen-32B","name":"DeepSeek: R1 Distill Qwen 32B","created":1738194830,"description":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on [Qwen 2.5 32B](https://huggingface.co/Qwen/Qwen2.5-32B), using outputs from [DeepSeek R1](/deepseek/deepseek-r1). It outperforms OpenAI's o1-mini across various benchmarks, achieving new state-of-the-art results for dense models.\\n\\nOther benchmark results include:\\n\\n- AIME 2024 pass@1: 72.6\\n- MATH-500 pass@1: 94.3\\n- CodeForces Rating: 1691\\n\\nThe model leverages fine-tuning from DeepSeek R1's outputs, enabling competitive performance comparable to larger frontier models.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen","instruct_type":"deepseek-r1"},"pricing":{"prompt":"0.00000029","completion":"0.00000029"},"top_provider":{"context_length":32768,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logprobs","max_tokens","presence_penalty","reasoning","response_format","stop","structured_outputs","temperature","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-07-31","expiration_date":null},{"id":"minimax/minimax-m2.1","canonical_slug":"minimax/minimax-m2.1","hugging_face_id":"MiniMaxAI/MiniMax-M2.1","name":"MiniMax: MiniMax M2.1","created":1766454997,"description":"MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development. With only 10 billion activated parameters, it delivers a major jump in real-world capability while maintaining exceptional latency, scalability, and cost efficiency.\n\nCompared to its predecessor, M2.1 delivers cleaner, more concise outputs and faster perceived response times. It shows leading multilingual coding performance across major systems and application languages, achieving 49.4% on Multi-SWE-Bench and 72.5% on SWE-Bench Multilingual, and serves as a versatile agent “brain” for IDEs, coding tools, and general-purpose assistance.\n\nTo avoid degrading this model's performance, MiniMax highly recommends preserving reasoning between turns. Learn more about using reasoning_details to pass back reasoning in our [docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#preserving-reasoning-blocks).","context_length":204800,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000029","completion":"0.00000095","input_cache_read":"0.00000003"},"top_provider":{"context_length":196608,"max_completion_tokens":196608,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":1,"top_p":0.9,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"x-ai/grok-3-mini","canonical_slug":"x-ai/grok-3-mini","hugging_face_id":"","name":"xAI: Grok 3 Mini","created":1749583245,"description":"A lightweight model that thinks before responding. Fast, smart, and great for logic-based tasks that do not require deep domain knowledge. The raw thinking traces are accessible.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.0000003","completion":"0.0000005","web_search":"0.005","input_cache_read":"0.000000075"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","logprobs","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-02-28","expiration_date":null},{"id":"z-ai/glm-4.6v","canonical_slug":"z-ai/glm-4.6-20251208","hugging_face_id":"zai-org/GLM-4.6V","name":"Z.ai: GLM 4.6V","created":1765207462,"description":"GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed media. It supports up to 128K tokens, processes complex page layouts and charts directly as visual inputs, and integrates native multimodal function calling to connect perception with downstream tool execution. The model also enables interleaved image-text generation and UI reconstruction workflows, including screenshot-to-HTML synthesis and iterative visual editing.","context_length":131072,"architecture":{"modality":"text+image+video->text","input_modalities":["image","text","video"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000003","completion":"0.0000009"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.8,"top_p":0.6,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"mistralai/codestral-2508","canonical_slug":"mistralai/codestral-2508","hugging_face_id":"","name":"Mistral: Codestral 2508","created":1754079630,"description":"Mistral's cutting-edge language model for coding released end of July 2025. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.\n\n[Blog Post](https://mistral.ai/news/codestral-25-08)","context_length":256000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000003","completion":"0.0000009","input_cache_read":"0.00000003"},"top_provider":{"context_length":256000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"tngtech/deepseek-r1t2-chimera","canonical_slug":"tngtech/deepseek-r1t2-chimera","hugging_face_id":"tngtech/DeepSeek-TNG-R1T2-Chimera","name":"TNG: DeepSeek R1T2 Chimera","created":1751986985,"description":"DeepSeek-TNG-R1T2-Chimera is the second-generation Chimera model from TNG Tech. It is a 671 B-parameter mixture-of-experts text-generation model assembled from DeepSeek-AI’s R1-0528, R1, and V3-0324 checkpoints with an Assembly-of-Experts merge. The tri-parent design yields strong reasoning performance while running roughly 20 % faster than the original R1 and more than 2× faster than R1-0528 under vLLM, giving a favorable cost-to-intelligence trade-off. The checkpoint supports contexts up to 60 k tokens in standard use (tested to ~130 k) and maintains consistent <think> token behaviour, making it suitable for long-context analysis, dialogue and other open-ended generation tasks.","context_length":163840,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":null},"pricing":{"prompt":"0.0000003","completion":"0.0000011","input_cache_read":"0.00000015"},"top_provider":{"context_length":163840,"max_completion_tokens":163840,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-07-31","expiration_date":null},{"id":"kwaipilot/kat-coder-pro-v2","canonical_slug":"kwaipilot/kat-coder-pro-v2-20260327","hugging_face_id":"","name":"Kwaipilot: KAT-Coder-Pro V2","created":1774649310,"description":"KAT-Coder-Pro V2 is the latest high-performance model in KwaiKAT’s KAT-Coder series, designed for complex enterprise-grade software engineering and SaaS integration. It builds on the agentic coding strengths of earlier versions, with a focus on large-scale production environments, multi-system coordination, and seamless integration across modern software stacks, while also supporting web aesthetics generation to produce production-grade landing pages and presentation decks.","context_length":256000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000003","completion":"0.0000012","input_cache_read":"0.00000006"},"top_provider":{"context_length":262144,"max_completion_tokens":144000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-2.5-flash-image","canonical_slug":"google/gemini-2.5-flash-image","hugging_face_id":"","name":"Google: Nano Banana (Gemini 2.5 Flash Image)","created":1759870431,"description":"Gemini 2.5 Flash Image, a.k.a. \"Nano Banana,\" is now generally available. It is a state of the art image generation model with contextual understanding. It is capable of image generation, edits, and multi-turn conversations. Aspect ratios can be controlled with the [image_config API Parameter](https://openrouter.ai/docs/features/multimodal/image-generation#image-aspect-ratio-configuration)","context_length":32768,"architecture":{"modality":"text+image->text+image","input_modalities":["image","text"],"output_modalities":["image","text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.0000003","completion":"0.0000025","image":"0.0000003","audio":"0.000001","internal_reasoning":"0.0000025","input_cache_read":"0.00000003","input_cache_write":"0.00000008333333333333334"},"top_provider":{"context_length":32768,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","stop","structured_outputs","temperature","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"google/gemini-2.5-flash","canonical_slug":"google/gemini-2.5-flash","hugging_face_id":"","name":"Google: Gemini 2.5 Flash","created":1750172488,"description":"Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in \"thinking\" capabilities, enabling it to provide responses with greater accuracy and nuanced context handling. \n\nAdditionally, Gemini 2.5 Flash is configurable through the \"max tokens for reasoning\" parameter, as described in the documentation (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning).","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["file","image","text","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.0000003","completion":"0.0000025","image":"0.0000003","audio":"0.000001","internal_reasoning":"0.0000025","input_cache_read":"0.00000003","input_cache_write":"0.00000008333333333333334"},"top_provider":{"context_length":1048576,"max_completion_tokens":65535,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"qwen/qwen-2.5-72b-instruct","canonical_slug":"qwen/qwen-2.5-72b-instruct","hugging_face_id":"Qwen/Qwen2.5-72B-Instruct","name":"Qwen2.5 72B Instruct","created":1726704000,"description":"Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2:\n\n- Significantly more knowledge and has greatly improved capabilities in coding and mathematics, thanks to our specialized expert models in these domains.\n\n- Significant improvements in instruction following, generating long texts (over 8K tokens), understanding structured data (e.g, tables), and generating structured outputs especially JSON. More resilient to the diversity of system prompts, enhancing role-play implementation and condition-setting for chatbots.\n\n- Long-context Support up to 128K tokens and can generate up to 8K tokens.\n\n- Multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.\n\nUsage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://huggingface.co/Qwen/Qwen1.5-110B-Chat/blob/main/LICENSE).","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen","instruct_type":"chatml"},"pricing":{"prompt":"0.00000038","completion":"0.0000004"},"top_provider":{"context_length":32000,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"moonshotai/kimi-k2.5","canonical_slug":"moonshotai/kimi-k2.5-0127","hugging_face_id":"moonshotai/Kimi-K2.5","name":"MoonshotAI: Kimi K2.5","created":1769487076,"description":"Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed visual and text tokens, it delivers strong performance in general reasoning, visual coding, and agentic tool-calling.","context_length":262144,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000003827","completion":"0.000001909","input_cache_read":"0.00000019135"},"top_provider":{"context_length":262144,"max_completion_tokens":65535,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","parallel_tool_calls","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"z-ai/glm-4.7","canonical_slug":"z-ai/glm-4.7-20251222","hugging_face_id":"zai-org/GLM-4.7","name":"Z.ai: GLM 4.7","created":1766378014,"description":"GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution. It demonstrates significant improvements in executing complex agent tasks while delivering more natural conversational experiences and superior front-end aesthetics.","context_length":200000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000039","completion":"0.00000175","input_cache_read":"0.000000195"},"top_provider":{"context_length":202752,"max_completion_tokens":65535,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"z-ai/glm-4.6","canonical_slug":"z-ai/glm-4.6","hugging_face_id":"zai-org/GLM-4.6","name":"Z.ai: GLM 4.6","created":1759235576,"description":"Compared with GLM-4.5, this generation brings several key improvements:\n\nLonger context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex agentic tasks.\nSuperior coding performance: The model achieves higher scores on code benchmarks and demonstrates better real-world performance in applications such as Claude Code、Cline、Roo Code and Kilo Code, including improvements in generating visually polished front-end pages.\nAdvanced reasoning: GLM-4.6 shows a clear improvement in reasoning performance and supports tool use during inference, leading to stronger overall capability.\nMore capable agents: GLM-4.6 exhibits stronger performance in tool using and search-based agents, and integrates more effectively within agent frameworks.\nRefined writing: Better aligns with human preferences in style and readability, and performs more naturally in role-playing scenarios.","context_length":200000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000039","completion":"0.0000019"},"top_provider":{"context_length":204800,"max_completion_tokens":204800,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"thedrummer/unslopnemo-12b","canonical_slug":"thedrummer/unslopnemo-12b","hugging_face_id":"TheDrummer/UnslopNemo-12B-v4.1","name":"TheDrummer: UnslopNemo 12B","created":1731103448,"description":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","context_length":32000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":"mistral"},"pricing":{"prompt":"0.0000004","completion":"0.0000004"},"top_provider":{"context_length":32768,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-04-30","expiration_date":null},{"id":"openai/gpt-4.1-mini","canonical_slug":"openai/gpt-4.1-mini-2025-04-14","hugging_face_id":"","name":"OpenAI: GPT-4.1 Mini","created":1744651381,"description":"GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost. It retains a 1 million token context window and scores 45.1% on hard instruction evals, 35.8% on MultiChallenge, and 84.1% on IFEval. Mini also shows strong coding ability (e.g., 31.6% on Aider’s polyglot diff benchmark) and vision understanding, making it suitable for interactive applications with tight performance constraints.","context_length":1047576,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000004","completion":"0.0000016","web_search":"0.01","input_cache_read":"0.0000001"},"top_provider":{"context_length":1047576,"max_completion_tokens":32768,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"mistralai/devstral-2512","canonical_slug":"mistralai/devstral-2512","hugging_face_id":"mistralai/Devstral-2-123B-Instruct-2512","name":"Mistral: Devstral 2 2512","created":1765285419,"description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window.\n\nDevstral 2 supports exploring codebases and orchestrating changes across multiple files while maintaining architecture-level context. It tracks framework dependencies, detects failures, and retries with corrections—solving challenges like bug fixing and modernizing legacy systems. The model can be fine-tuned to prioritize specific languages or optimize for large enterprise codebases. It is available under a modified MIT license.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000004","completion":"0.000002","input_cache_read":"0.00000004"},"top_provider":{"context_length":262144,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"mistralai/devstral-medium","canonical_slug":"mistralai/devstral-medium-2507","hugging_face_id":"","name":"Mistral: Devstral Medium","created":1752161321,"description":"Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI. Positioned as a step up from Devstral Small, it achieves 61.6% on SWE-Bench Verified, placing it ahead of Gemini 2.5 Pro and GPT-4.1 in code-related tasks, at a fraction of the cost. It is designed for generalization across prompt styles and tool use in code agents and frameworks.\n\nDevstral Medium is available via API only (not open-weight), and supports enterprise deployment on private infrastructure, with optional fine-tuning capabilities.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000004","completion":"0.000002","input_cache_read":"0.00000004"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"mistralai/mistral-medium-3","canonical_slug":"mistralai/mistral-medium-3","hugging_face_id":"","name":"Mistral: Mistral Medium 3","created":1746627341,"description":"Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost compared to traditional large models, making it suitable for scalable deployments across professional and industrial use cases.\n\nThe model excels in domains such as coding, STEM reasoning, and enterprise adaptation. It supports hybrid, on-prem, and in-VPC deployments and is optimized for integration into custom workflows. Mistral Medium 3 offers competitive accuracy relative to larger models like Claude Sonnet 3.5/3.7, Llama 4 Maverick, and Command R+, while maintaining broad compatibility across cloud environments.","context_length":131072,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000004","completion":"0.000002","input_cache_read":"0.00000004"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"moonshotai/kimi-k2-0905","canonical_slug":"moonshotai/kimi-k2-0905","hugging_face_id":"moonshotai/Kimi-K2-Instruct-0905","name":"MoonshotAI: Kimi K2 0905","created":1757021147,"description":"Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2). It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It supports long-context inference up to 256k tokens, extended from the previous 128k.\n\nThis update improves agentic coding with higher accuracy and better generalization across scaffolds, and enhances frontend coding with more aesthetic and functional outputs for web, 3D, and related tasks. Kimi K2 is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. It excels across coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) benchmarks. The model is trained with a novel stack incorporating the MuonClip optimizer for stable large-scale MoE training.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000004","completion":"0.000002"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-12-31","expiration_date":null},{"id":"mistralai/mistral-medium-3.1","canonical_slug":"mistralai/mistral-medium-3.1","hugging_face_id":"","name":"Mistral: Mistral Medium 3.1","created":1755095639,"description":"Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost compared to traditional large models, making it suitable for scalable deployments across professional and industrial use cases.\n\nThe model excels in domains such as coding, STEM reasoning, and enterprise adaptation. It supports hybrid, on-prem, and in-VPC deployments and is optimized for integration into custom workflows. Mistral Medium 3.1 offers competitive accuracy relative to larger models like Claude Sonnet 3.5/3.7, Llama 4 Maverick, and Command R+, while maintaining broad compatibility across cloud environments.","context_length":131072,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000004","completion":"0.000002","input_cache_read":"0.00000004"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"undi95/remm-slerp-l2-13b","canonical_slug":"undi95/remm-slerp-l2-13b","hugging_face_id":"Undi95/ReMM-SLERP-L2-13B","name":"ReMM SLERP 13B","created":1689984000,"description":"A recreation trial of the original MythoMax-L2-B13 but with updated models. #merge","context_length":4096,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama2","instruct_type":"alpaca"},"pricing":{"prompt":"0.00000045","completion":"0.00000065"},"top_provider":{"context_length":6144,"max_completion_tokens":4096,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_a","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-06-30","expiration_date":null},{"id":"deepseek/deepseek-r1-0528","canonical_slug":"deepseek/deepseek-r1-0528","hugging_face_id":"deepseek-ai/DeepSeek-R1-0528","name":"DeepSeek: R1 0528","created":1748455170,"description":"May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass.\n\nFully open-source model.","context_length":163840,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":"deepseek-r1"},"pricing":{"prompt":"0.00000045","completion":"0.00000215","input_cache_read":"0.000000225"},"top_provider":{"context_length":163840,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"qwen/qwen3-vl-235b-a22b-thinking","canonical_slug":"qwen/qwen3-vl-235b-a22b-thinking","hugging_face_id":"Qwen/Qwen3-VL-235B-A22B-Thinking","name":"Qwen: Qwen3 VL 235B A22B Thinking","created":1758668690,"description":"Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video. The Thinking model is optimized for multimodal reasoning in STEM and math. The series emphasizes robust perception (recognition of diverse real-world and synthetic categories), spatial understanding (2D/3D grounding), and long-form visual comprehension, with competitive results on public multimodal benchmarks for both perception and reasoning.\n\nBeyond analysis, Qwen3-VL supports agentic interaction and tool use: it can follow complex instructions over multi-image, multi-turn dialogues; align text to video timelines for precise temporal queries; and operate GUI elements for automation tasks. The models also enable visual coding workflows, turning sketches or mockups into code and assisting with UI debugging, while maintaining strong text-only performance comparable to the flagship Qwen3 language models. This makes Qwen3-VL suitable for production scenarios spanning document AI, multilingual OCR, software/UI assistance, spatial/embodied tasks, and research on vision-language agents.","context_length":131072,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000045","completion":"0.0000035"},"top_provider":{"context_length":262144,"max_completion_tokens":262144,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.8,"top_p":0.95,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":1},"knowledge_cutoff":"2025-03-31","expiration_date":null},{"id":"mistralai/mistral-large-2512","canonical_slug":"mistralai/mistral-large-2512","hugging_face_id":"","name":"Mistral: Mistral Large 3 2512","created":1764624472,"description":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","context_length":256000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.0000005","completion":"0.0000015","input_cache_read":"0.00000005"},"top_provider":{"context_length":262144,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.0645,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-3.5-turbo","canonical_slug":"openai/gpt-3.5-turbo","hugging_face_id":null,"name":"OpenAI: GPT-3.5 Turbo","created":1685232000,"description":"GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks.\n\nTraining data up to Sep 2021.","context_length":16385,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000005","completion":"0.0000015"},"top_provider":{"context_length":16385,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2021-09-30","expiration_date":null},{"id":"google/gemini-3-flash-preview","canonical_slug":"google/gemini-3-flash-preview-20251217","hugging_face_id":"","name":"Google: Gemini 3 Flash Preview","created":1765987078,"description":"Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance. It delivers near Pro level reasoning and tool use performance with substantially lower latency than larger Gemini variants, making it well suited for interactive development, long running agent loops, and collaborative coding tasks. Compared to Gemini 2.5 Flash, it provides broad quality improvements across reasoning, multimodal understanding, and reliability.\n\nThe model supports a 1M token context window and multimodal inputs including text, images, audio, video, and PDFs, with text output. It includes configurable reasoning via thinking levels (minimal, low, medium, high), structured output, tool use, and automatic context caching. Gemini 3 Flash Preview is optimized for users who want strong reasoning and agentic behavior without the cost or latency of full scale frontier models.","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.0000005","completion":"0.000003","image":"0.0000005","audio":"0.000001","internal_reasoning":"0.000003","input_cache_read":"0.00000005","input_cache_write":"0.00000008333333333333334"},"top_provider":{"context_length":1048576,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-3.1-flash-image-preview","canonical_slug":"google/gemini-3.1-flash-image-preview-20260226","hugging_face_id":"","name":"Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)","created":1772119558,"description":"Gemini 3.1 Flash Image Preview, a.k.a. \"Nano Banana 2,\" is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines advanced contextual understanding with fast, cost-efficient inference, making complex image generation and iterative edits significantly more accessible. Aspect ratios can be controlled with the [image_config API Parameter](https://openrouter.ai/docs/features/multimodal/image-generation#image-aspect-ratio-configuration)","context_length":131072,"architecture":{"modality":"text+image->text+image","input_modalities":["image","text"],"output_modalities":["image","text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.0000005","completion":"0.000003"},"top_provider":{"context_length":65536,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"meta-llama/llama-3-70b-instruct","canonical_slug":"meta-llama/llama-3-70b-instruct","hugging_face_id":"meta-llama/Meta-Llama-3-70B-Instruct","name":"Meta: Llama 3 70B Instruct","created":1713398400,"description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors. This 70B instruct-tuned version was optimized for high quality dialogue usecases.\n\nIt has demonstrated strong performance compared to leading closed-source models in human evaluations.\n\nTo read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/).","context_length":8192,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"llama3"},"pricing":{"prompt":"0.00000051","completion":"0.00000074"},"top_provider":{"context_length":8192,"max_completion_tokens":8000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"qwen/qwen3.5-397b-a17b","canonical_slug":"qwen/qwen3.5-397b-a17b-20260216","hugging_face_id":"Qwen/Qwen3.5-397B-A17B","name":"Qwen: Qwen3.5 397B A17B","created":1771223018,"description":"The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers state-of-the-art performance comparable to leading-edge models across a wide range of tasks, including language understanding, logical reasoning, code generation, agent-based tasks, image understanding, video understanding, and graphical user interface (GUI) interactions. With its robust code-generation and agent capabilities, the model exhibits strong generalization across diverse agent.","context_length":256000,"architecture":{"modality":"text+image+video->text","input_modalities":["text","image","video"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000055","completion":"0.0000035","input_cache_read":"0.00000055"},"top_provider":{"context_length":262144,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.6,"top_p":0.95,"top_k":20,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"z-ai/glm-4.5v","canonical_slug":"z-ai/glm-4.5v","hugging_face_id":"zai-org/GLM-4.5V","name":"Z.ai: GLM 4.5V","created":1754922288,"description":"GLM-4.5V is a vision-language foundation model for multimodal agent applications. Built on a Mixture-of-Experts (MoE) architecture with 106B parameters and 12B activated parameters, it achieves state-of-the-art results in video understanding, image Q&A, OCR, and document parsing, with strong gains in front-end web coding, grounding, and spatial reasoning. It offers a hybrid inference mode: a \"thinking mode\" for deep reasoning and a \"non-thinking mode\" for fast responses. Reasoning behavior can be toggled via the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)","context_length":65536,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000006","completion":"0.0000018","input_cache_read":"0.00000011"},"top_provider":{"context_length":65536,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.75,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-12-31","expiration_date":null},{"id":"nvidia/llama-3.1-nemotron-ultra-253b-v1","canonical_slug":"nvidia/llama-3.1-nemotron-ultra-253b-v1","hugging_face_id":"nvidia/Llama-3_1-Nemotron-Ultra-253B-v1","name":"NVIDIA: Llama 3.1 Nemotron Ultra 253B v1","created":1744115059,"description":"Llama-3.1-Nemotron-Ultra-253B-v1 is a large language model (LLM) optimized for advanced reasoning, human-interactive chat, retrieval-augmented generation (RAG), and tool-calling tasks. Derived from Meta’s Llama-3.1-405B-Instruct, it has been significantly customized using Neural Architecture Search (NAS), resulting in enhanced efficiency, reduced memory usage, and improved inference latency. The model supports a context length of up to 128K tokens and can operate efficiently on an 8x NVIDIA H100 node.\n\nNote: you must include `detailed thinking on` in the system prompt to enable reasoning. Please see [Usage Recommendations](https://huggingface.co/nvidia/Llama-3_1-Nemotron-Ultra-253B-v1#quick-start-and-usage-recommendations) for more.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":null},"pricing":{"prompt":"0.0000006","completion":"0.0000018"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","structured_outputs","temperature","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2024-03-31","expiration_date":null},{"id":"z-ai/glm-4.5","canonical_slug":"z-ai/glm-4.5","hugging_face_id":"zai-org/GLM-4.5","name":"Z.ai: GLM 4.5","created":1753471347,"description":"GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications. It leverages a Mixture-of-Experts (MoE) architecture and supports a context length of up to 128k tokens. GLM-4.5 delivers significantly enhanced capabilities in reasoning, code generation, and agent alignment. It supports a hybrid inference mode with two options, a \"thinking mode\" designed for complex reasoning and tool use, and a \"non-thinking mode\" optimized for instant responses. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000006","completion":"0.0000022"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":0.75,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-12-31","expiration_date":null},{"id":"openai/gpt-audio-mini","canonical_slug":"openai/gpt-audio-mini","hugging_face_id":"","name":"OpenAI: GPT Audio Mini","created":1768859419,"description":"A cost-efficient version of GPT Audio. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Input is priced at $0.60 per million tokens and output is priced at $2.40 per million tokens.","context_length":128000,"architecture":{"modality":"text+audio->text+audio","input_modalities":["text","audio"],"output_modalities":["text","audio"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000006","completion":"0.0000024","audio":"0.0000006"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"moonshotai/kimi-k2-thinking","canonical_slug":"moonshotai/kimi-k2-thinking-20251106","hugging_face_id":"moonshotai/Kimi-K2-Thinking","name":"MoonshotAI: Kimi K2 Thinking","created":1762440622,"description":"Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in Kimi K2, it activates 32 billion parameters per forward pass and supports 256 k-token context windows. The model is optimized for persistent step-by-step thought, dynamic tool invocation, and complex reasoning workflows that span hundreds of turns. It interleaves step-by-step reasoning with tool use, enabling autonomous research, coding, and writing that can persist for hundreds of sequential actions without drift.\n\nIt sets new open-source benchmarks on HLE, BrowseComp, SWE-Multilingual, and LiveCodeBench, while maintaining stable multi-agent behavior through 200–300 tool calls. Built on a large-scale MoE architecture with MuonClip optimization, it combines strong reasoning depth with high inference efficiency for demanding agentic and analytical tasks.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000006","completion":"0.0000025"},"top_provider":{"context_length":262144,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemma-2-27b-it","canonical_slug":"google/gemma-2-27b-it","hugging_face_id":"google/gemma-2-27b-it","name":"Google: Gemma 2 27B","created":1720828800,"description":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini).\n\nGemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning.\n\nSee the [launch announcement](https://blog.google/technology/developers/google-gemma-2/) for more details. Usage of Gemma is subject to Google's [Gemma Terms of Use](https://ai.google.dev/gemma/terms).","context_length":8192,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":"gemma"},"pricing":{"prompt":"0.00000065","completion":"0.00000065"},"top_provider":{"context_length":8192,"max_completion_tokens":2048,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","stop","structured_outputs","temperature","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"sao10k/l3.3-euryale-70b","canonical_slug":"sao10k/l3.3-euryale-70b-v2.3","hugging_face_id":"Sao10K/L3.3-70B-Euryale-v2.3","name":"Sao10K: Llama 3.3 Euryale 70B","created":1734535928,"description":"Euryale L3.3 70B is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k). It is the successor of [Euryale L3 70B v2.2](/models/sao10k/l3-euryale-70b).","context_length":8192,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"llama3"},"pricing":{"prompt":"0.00000065","completion":"0.00000075"},"top_provider":{"context_length":131072,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"qwen/qwen3-coder-plus","canonical_slug":"qwen/qwen3-coder-plus","hugging_face_id":"","name":"Qwen: Qwen3 Coder Plus","created":1758662707,"description":"Qwen3 Coder Plus is Alibaba's proprietary version of the Open Source Qwen3 Coder 480B A35B. It is a powerful coding agent model specializing in autonomous programming via tool calling and environment interaction, combining coding proficiency with versatile general-purpose abilities.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen3","instruct_type":null},"pricing":{"prompt":"0.00000065","completion":"0.00000325","input_cache_read":"0.00000013"},"top_provider":{"context_length":1000000,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["max_tokens","presence_penalty","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-06-30","expiration_date":null},{"id":"z-ai/glm-5","canonical_slug":"z-ai/glm-5-20260211","hugging_face_id":"zai-org/GLM-5","name":"Z.ai: GLM 5","created":1770829182,"description":"GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows. Built for expert developers, it delivers production-grade performance on large-scale programming tasks, rivaling leading closed-source models. With advanced agentic planning, deep backend reasoning, and iterative self-correction, GLM-5 moves beyond code generation to full-system construction and autonomous execution.","context_length":202752,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000072","completion":"0.0000023"},"top_provider":{"context_length":80000,"max_completion_tokens":131072,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_logprobs","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.4-mini","canonical_slug":"openai/gpt-5.4-mini-20260317","hugging_face_id":"","name":"OpenAI: GPT-5.4 Mini","created":1773748178,"description":"GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads. It supports text and image inputs with strong performance across reasoning, coding, and tool use, while reducing latency and cost for large-scale deployments.\n\nThe model is designed for production environments that require a balance of capability and efficiency, making it well suited for chat applications, coding assistants, and agent workflows that operate at scale. GPT-5.4 mini delivers reliable instruction following, solid multi-step reasoning, and consistent performance across diverse tasks with improved cost efficiency.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000075","completion":"0.0000045","web_search":"0.01","input_cache_read":"0.000000075"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-08-31","expiration_date":null},{"id":"qwen/qwen3-max-thinking","canonical_slug":"qwen/qwen3-max-thinking-20260123","hugging_face_id":null,"name":"Qwen: Qwen3 Max Thinking","created":1770671901,"description":"Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning. By significantly scaling model capacity and reinforcement learning compute, it delivers major gains in factual accuracy, complex reasoning, instruction following, alignment with human preferences, and agentic behavior.","context_length":262144,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Qwen","instruct_type":null},"pricing":{"prompt":"0.00000078","completion":"0.0000039"},"top_provider":{"context_length":262144,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","presence_penalty","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"deepseek/deepseek-r1-distill-llama-70b","canonical_slug":"deepseek/deepseek-r1-distill-llama-70b","hugging_face_id":"deepseek-ai/DeepSeek-R1-Distill-Llama-70B","name":"DeepSeek: R1 Distill Llama 70B","created":1737663169,"description":"DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1). The model combines advanced distillation techniques to achieve high performance across multiple benchmarks, including:\n\n- AIME 2024 pass@1: 70.0\n- MATH-500 pass@1: 94.5\n- CodeForces Rating: 1633\n\nThe model leverages fine-tuning from DeepSeek R1's outputs, enabling competitive performance comparable to larger frontier models.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"deepseek-r1"},"pricing":{"prompt":"0.0000008","completion":"0.0000008"},"top_provider":{"context_length":8192,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-07-31","expiration_date":null},{"id":"qwen/qwen2.5-vl-72b-instruct","canonical_slug":"qwen/qwen2.5-vl-72b-instruct","hugging_face_id":"Qwen/Qwen2.5-VL-72B-Instruct","name":"Qwen: Qwen2.5 VL 72B Instruct","created":1738410311,"description":"Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.","context_length":131072,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Qwen","instruct_type":null},"pricing":{"prompt":"0.0000008","completion":"0.000001"},"top_provider":{"context_length":128000,"max_completion_tokens":128000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","max_tokens","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"openai/gpt-3.5-turbo-0613","canonical_slug":"openai/gpt-3.5-turbo-0613","hugging_face_id":null,"name":"OpenAI: GPT-3.5 Turbo (older v0613)","created":1706140800,"description":"GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks.\n\nTraining data up to Sep 2021.","context_length":4095,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000001","completion":"0.000002"},"top_provider":{"context_length":4095,"max_completion_tokens":4096,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2021-09-30","expiration_date":null},{"id":"anthropic/claude-haiku-4.5","canonical_slug":"anthropic/claude-4.5-haiku-20251001","hugging_face_id":"","name":"Anthropic: Claude Haiku 4.5","created":1760547638,"description":"Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance across reasoning, coding, and computer-use tasks, Haiku 4.5 brings frontier-level capability to real-time and high-volume applications.\n\nIt introduces extended thinking to the Haiku line; enabling controllable reasoning depth, summarized or interleaved thought output, and tool-assisted workflows with full support for coding, bash, web search, and computer-use tools. Scoring >73% on SWE-bench Verified, Haiku 4.5 ranks among the world’s best coding models while maintaining exceptional responsiveness for sub-agents, parallelized execution, and scaled deployment.","context_length":200000,"architecture":{"modality":"text+image->text","input_modalities":["image","text"],"output_modalities":["text"],"tokenizer":"Claude","instruct_type":null},"pricing":{"prompt":"0.000001","completion":"0.000005","web_search":"0.01","input_cache_read":"0.0000001","input_cache_write":"0.00000125"},"top_provider":{"context_length":200000,"max_completion_tokens":64000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/o4-mini-high","canonical_slug":"openai/o4-mini-high-2025-04-16","hugging_face_id":"","name":"OpenAI: o4 Mini High","created":1744824212,"description":"OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high. \n\nOpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning and coding performance across benchmarks like AIME (99.5% with Python) and SWE-bench, outperforming its predecessor o3-mini and even approaching o3 in some domains.\n\nDespite its smaller size, o4-mini exhibits high accuracy in STEM tasks, visual problem solving (e.g., MathVista, MMMU), and code editing. It is especially well-suited for high-throughput scenarios where latency or cost is critical. Thanks to its efficient architecture and refined reinforcement learning training, o4-mini can chain tools, generate structured outputs, and solve multi-step tasks with minimal delay—often in under a minute.","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000011","completion":"0.0000044","web_search":"0.01","input_cache_read":"0.000000275"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"openai/o3-mini-high","canonical_slug":"openai/o3-mini-high-2025-01-31","hugging_face_id":"","name":"OpenAI: o3 Mini High","created":1739372611,"description":"OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high. \n\no3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding. The model features three adjustable reasoning effort levels and supports key developer capabilities including function calling, structured outputs, and streaming, though it does not include vision processing capabilities.\n\nThe model demonstrates significant improvements over its predecessor, with expert testers preferring its responses 56% of the time and noting a 39% reduction in major errors on complex questions. With medium reasoning effort settings, o3-mini matches the performance of the larger o1 model on challenging reasoning evaluations like AIME and GPQA, while maintaining lower latency and cost.","context_length":200000,"architecture":{"modality":"text+file->text","input_modalities":["text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000011","completion":"0.0000044","input_cache_read":"0.00000055"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/o3-mini","canonical_slug":"openai/o3-mini-2025-01-31","hugging_face_id":"","name":"OpenAI: o3 Mini","created":1738351721,"description":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding.\n\nThis model supports the `reasoning_effort` parameter, which can be set to \"high\", \"medium\", or \"low\" to control the thinking time of the model. The default is \"medium\". OpenRouter also offers the model slug `openai/o3-mini-high` to default the parameter to \"high\".\n\nThe model features three adjustable reasoning effort levels and supports key developer capabilities including function calling, structured outputs, and streaming, though it does not include vision processing capabilities.\n\nThe model demonstrates significant improvements over its predecessor, with expert testers preferring its responses 56% of the time and noting a 39% reduction in major errors on complex questions. With medium reasoning effort settings, o3-mini matches the performance of the larger o1 model on challenging reasoning evaluations like AIME and GPQA, while maintaining lower latency and cost.","context_length":200000,"architecture":{"modality":"text+file->text","input_modalities":["text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000011","completion":"0.0000044","input_cache_read":"0.00000055"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/o4-mini","canonical_slug":"openai/o4-mini-2025-04-16","hugging_face_id":"","name":"OpenAI: o4 Mini","created":1744820942,"description":"OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning and coding performance across benchmarks like AIME (99.5% with Python) and SWE-bench, outperforming its predecessor o3-mini and even approaching o3 in some domains.\n\nDespite its smaller size, o4-mini exhibits high accuracy in STEM tasks, visual problem solving (e.g., MathVista, MMMU), and code editing. It is especially well-suited for high-throughput scenarios where latency or cost is critical. Thanks to its efficient architecture and refined reinforcement learning training, o4-mini can chain tools, generate structured outputs, and solve multi-step tasks with minimal delay—often in under a minute.","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000011","completion":"0.0000044","web_search":"0.01","input_cache_read":"0.000000275"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"deepcogito/cogito-v2.1-671b","canonical_slug":"deepcogito/cogito-v2.1-671b-20251118","hugging_face_id":"","name":"Deep Cogito: Cogito v2.1 671B","created":1763071233,"description":"Cogito v2.1 671B MoE represents one of the strongest open models globally, matching performance of frontier closed and open models. This model is trained using self play with reinforcement learning to reach state-of-the-art performance on multiple categories (instruction following, coding, longer queries and creative writing). This advanced system demonstrates significant progress toward scalable superintelligence through policy improvement.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00000125"},"top_provider":{"context_length":128000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5-chat","canonical_slug":"openai/gpt-5-chat-2025-08-07","hugging_face_id":"","name":"OpenAI: GPT-5 Chat","created":1754587837,"description":"GPT-5 Chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications.","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","web_search":"0.01","input_cache_read":"0.000000125"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs"],"default_parameters":{},"knowledge_cutoff":"2024-09-30","expiration_date":null},{"id":"openai/gpt-5","canonical_slug":"openai/gpt-5-2025-08-07","hugging_face_id":"","name":"OpenAI: GPT-5","created":1754587413,"description":"GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. It supports test-time routing features and advanced prompt understanding, including user-specified intent like \"think hard about this.\" Improvements include reductions in hallucination, sycophancy, and better performance in coding, writing, and health-related tasks.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","web_search":"0.01","input_cache_read":"0.000000125"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2024-09-30","expiration_date":null},{"id":"openai/gpt-5-codex","canonical_slug":"openai/gpt-5-codex","hugging_face_id":"","name":"OpenAI: GPT-5 Codex","created":1758643403,"description":"GPT-5-Codex is a specialized version of GPT-5 optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks. The model supports building projects from scratch, feature development, debugging, large-scale refactoring, and code review. Compared to GPT-5, Codex is more steerable, adheres closely to developer instructions, and produces cleaner, higher-quality code outputs. Reasoning effort can be adjusted with the `reasoning.effort` parameter. Read the [docs here](https://openrouter.ai/docs/use-cases/reasoning-tokens#reasoning-effort-level)\n\nCodex integrates into developer environments including the CLI, IDE extensions, GitHub, and cloud tasks. It adapts reasoning effort dynamically—providing fast responses for small tasks while sustaining extended multi-hour runs for large projects. The model is trained to perform structured code reviews, catching critical flaws by reasoning over dependencies and validating behavior against tests. It also supports multimodal inputs such as images or screenshots for UI development and integrates tool use for search, dependency installation, and environment setup. Codex is intended specifically for agentic coding applications.","context_length":400000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","input_cache_read":"0.000000125"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-09-30","expiration_date":null},{"id":"openai/gpt-5.1-codex","canonical_slug":"openai/gpt-5.1-codex-20251113","hugging_face_id":"","name":"OpenAI: GPT-5.1-Codex","created":1763060298,"description":"GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks. The model supports building projects from scratch, feature development, debugging, large-scale refactoring, and code review. Compared to GPT-5.1, Codex is more steerable, adheres closely to developer instructions, and produces cleaner, higher-quality code outputs. Reasoning effort can be adjusted with the `reasoning.effort` parameter. Read the [docs here](https://openrouter.ai/docs/use-cases/reasoning-tokens#reasoning-effort-level)\n\nCodex integrates into developer environments including the CLI, IDE extensions, GitHub, and cloud tasks. It adapts reasoning effort dynamically—providing fast responses for small tasks while sustaining extended multi-hour runs for large projects. The model is trained to perform structured code reviews, catching critical flaws by reasoning over dependencies and validating behavior against tests. It also supports multimodal inputs such as images or screenshots for UI development and integrates tool use for search, dependency installation, and environment setup. Codex is intended specifically for agentic coding applications.","context_length":400000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","input_cache_read":"0.000000125"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.1-chat","canonical_slug":"openai/gpt-5.1-chat-20251113","hugging_face_id":"","name":"OpenAI: GPT-5.1 Chat","created":1763060302,"description":"GPT-5.1 Chat (AKA Instant is the fast, lightweight member of the 5.1 family, optimized for low-latency chat while retaining strong general intelligence. It uses adaptive reasoning to selectively “think” on harder queries, improving accuracy on math, coding, and multi-step tasks without slowing down typical conversations. The model is warmer and more conversational by default, with better instruction following and more stable short-form reasoning. GPT-5.1 Chat is designed for high-throughput, interactive workloads where responsiveness and consistency matter more than deep deliberation.\n","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","web_search":"0.01","input_cache_read":"0.000000125"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.1-codex-max","canonical_slug":"openai/gpt-5.1-codex-max-20251204","hugging_face_id":"","name":"OpenAI: GPT-5.1-Codex-Max","created":1764878934,"description":"GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks. It is based on an updated version of the 5.1 reasoning stack and trained on agentic workflows spanning software engineering, mathematics, and research. \nGPT-5.1-Codex-Max delivers faster performance, improved reasoning, and higher token efficiency across the development lifecycle. ","context_length":400000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","web_search":"0.01","input_cache_read":"0.000000125"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.1","canonical_slug":"openai/gpt-5.1-20251113","hugging_face_id":"","name":"OpenAI: GPT-5.1","created":1763060305,"description":"GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning to allocate computation dynamically, responding quickly to simple queries while spending more depth on complex tasks. The model produces clearer, more grounded explanations with reduced jargon, making it easier to follow even on technical or multi-step problems.\n\nBuilt for broad task coverage, GPT-5.1 delivers consistent gains across math, coding, and structured analysis workloads, with more coherent long-form answers and improved tool-use reliability. It also features refined conversational alignment, enabling warmer, more intuitive responses without compromising precision. GPT-5.1 serves as the primary full-capability successor to GPT-5","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","web_search":"0.01","input_cache_read":"0.000000125"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-2.5-pro-preview","canonical_slug":"google/gemini-2.5-pro-preview-06-05","hugging_face_id":"","name":"Google: Gemini 2.5 Pro Preview 06-05","created":1749137257,"description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy and nuanced context handling. Gemini 2.5 Pro achieves top-tier performance on multiple benchmarks, including first-place positioning on the LMArena leaderboard, reflecting superior human-preference alignment and complex problem-solving abilities.\n","context_length":1048576,"architecture":{"modality":"text+image+file+audio->text","input_modalities":["file","image","text","audio"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","image":"0.00000125","audio":"0.00000125","internal_reasoning":"0.00001","input_cache_read":"0.000000125","input_cache_write":"0.000000375"},"top_provider":{"context_length":1048576,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"google/gemini-2.5-pro","canonical_slug":"google/gemini-2.5-pro","hugging_face_id":"","name":"Google: Gemini 2.5 Pro","created":1750169544,"description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy and nuanced context handling. Gemini 2.5 Pro achieves top-tier performance on multiple benchmarks, including first-place positioning on the LMArena leaderboard, reflecting superior human-preference alignment and complex problem-solving abilities.","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","image":"0.00000125","audio":"0.00000125","internal_reasoning":"0.00001","input_cache_read":"0.000000125","input_cache_write":"0.000000375"},"top_provider":{"context_length":1048576,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"google/gemini-2.5-pro-preview-05-06","canonical_slug":"google/gemini-2.5-pro-preview-03-25","hugging_face_id":"","name":"Google: Gemini 2.5 Pro Preview 05-06","created":1746578513,"description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy and nuanced context handling. Gemini 2.5 Pro achieves top-tier performance on multiple benchmarks, including first-place positioning on the LMArena leaderboard, reflecting superior human-preference alignment and complex problem-solving abilities.","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","image":"0.00000125","audio":"0.00000125","internal_reasoning":"0.00001","input_cache_read":"0.000000125","input_cache_write":"0.000000375"},"top_provider":{"context_length":1048576,"max_completion_tokens":65535,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"sao10k/l3.1-euryale-70b","canonical_slug":"sao10k/l3.1-euryale-70b","hugging_face_id":"Sao10K/L3.1-70B-Euryale-v2.2","name":"Sao10K: Llama 3.1 Euryale 70B v2.2","created":1724803200,"description":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k). It is the successor of [Euryale L3 70B v2.1](/models/sao10k/l3-euryale-70b).","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Llama3","instruct_type":"llama3"},"pricing":{"prompt":"0.00000148","completion":"0.00000148"},"top_provider":{"context_length":8192,"max_completion_tokens":8192,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","min_p","presence_penalty","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"openai/gpt-3.5-turbo-instruct","canonical_slug":"openai/gpt-3.5-turbo-instruct","hugging_face_id":null,"name":"OpenAI: GPT-3.5 Turbo Instruct","created":1695859200,"description":"This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations. Training data: up to Sep 2021.","context_length":4095,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":"chatml"},"pricing":{"prompt":"0.0000015","completion":"0.000002"},"top_provider":{"context_length":4095,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2021-09-30","expiration_date":null},{"id":"openai/gpt-5.3-codex","canonical_slug":"openai/gpt-5.3-codex-20260224","hugging_face_id":"","name":"OpenAI: GPT-5.3-Codex","created":1771959164,"description":"GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2. It achieves state-of-the-art results on SWE-Bench Pro and strong performance on Terminal-Bench 2.0 and OSWorld-Verified, reflecting improved multi-language coding, terminal proficiency, and real-world computer-use skills. The model is optimized for long-running, tool-using workflows and supports interactive steering during execution, making it suitable for complex development tasks, debugging, deployment, and iterative product work.\n\nBeyond coding, GPT-5.3-Codex performs strongly on structured knowledge-work benchmarks such as GDPval, supporting tasks like document drafting, spreadsheet analysis, slide creation, and operational research across domains. It is trained with enhanced cybersecurity awareness, including vulnerability identification capabilities, and deployed with additional safeguards for high-risk use cases. Compared to prior Codex models, it is more token-efficient and approximately 25% faster, targeting professional end-to-end workflows that span reasoning, execution, and computer interaction.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000175","completion":"0.000014","web_search":"0.01","input_cache_read":"0.000000175"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.3-chat","canonical_slug":"openai/gpt-5.3-chat-20260303","hugging_face_id":"","name":"OpenAI: GPT-5.3 Chat","created":1772564061,"description":"GPT-5.3 Chat is an update to ChatGPT's most-used model that makes everyday conversations smoother, more useful, and more directly helpful. It delivers more accurate answers with better contextualization and significantly reduces unnecessary refusals, caveats, and overly cautious phrasing that can interrupt conversational flow.","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000175","completion":"0.000014","web_search":"0.1","input_cache_read":"0.000000175"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.2","canonical_slug":"openai/gpt-5.2-20251211","hugging_face_id":"","name":"OpenAI: GPT-5.2","created":1765389775,"description":"GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1. It uses adaptive reasoning to allocate computation dynamically, responding quickly to simple queries while spending more depth on complex tasks.\n\nBuilt for broad task coverage, GPT-5.2 delivers consistent gains across math, coding, sciende, and tool calling workloads, with more coherent long-form answers and improved tool-use reliability.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000175","completion":"0.000014","web_search":"0.01","input_cache_read":"0.000000175"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.2-chat","canonical_slug":"openai/gpt-5.2-chat-20251211","hugging_face_id":"","name":"OpenAI: GPT-5.2 Chat","created":1765389783,"description":"GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligence. It uses adaptive reasoning to selectively “think” on harder queries, improving accuracy on math, coding, and multi-step tasks without slowing down typical conversations. The model is warmer and more conversational by default, with better instruction following and more stable short-form reasoning. GPT-5.2 Chat is designed for high-throughput, interactive workloads where responsiveness and consistency matter more than deep deliberation.","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000175","completion":"0.000014","web_search":"0.01","input_cache_read":"0.000000175"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5.2-codex","canonical_slug":"openai/gpt-5.2-codex-20260114","hugging_face_id":"","name":"OpenAI: GPT-5.2-Codex","created":1768409315,"description":"GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks. The model supports building projects from scratch, feature development, debugging, large-scale refactoring, and code review. Compared to GPT-5.1-Codex, 5.2-Codex is more steerable, adheres closely to developer instructions, and produces cleaner, higher-quality code outputs. Reasoning effort can be adjusted with the `reasoning.effort` parameter. Read the [docs here](https://openrouter.ai/docs/use-cases/reasoning-tokens#reasoning-effort-level)\n\nCodex integrates into developer environments including the CLI, IDE extensions, GitHub, and cloud tasks. It adapts reasoning effort dynamically—providing fast responses for small tasks while sustaining extended multi-hour runs for large projects. The model is trained to perform structured code reviews, catching critical flaws by reasoning over dependencies and validating behavior against tests. It also supports multimodal inputs such as images or screenshots for UI development and integrates tool use for search, dependency installation, and environment setup. Codex is intended specifically for agentic coding applications.","context_length":400000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000175","completion":"0.000014","web_search":"0.01","input_cache_read":"0.000000175"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"mistralai/mistral-large-2411","canonical_slug":"mistralai/mistral-large-2411","hugging_face_id":"","name":"Mistral Large 2411","created":1731978685,"description":"Mistral Large 2 2411 is an update of [Mistral Large 2](/mistralai/mistral-large) released together with [Pixtral Large 2411](/mistralai/pixtral-large-2411)\n\nIt provides a significant upgrade on the previous [Mistral Large 24.07](/mistralai/mistral-large-2407), with notable improvements in long context understanding, a new system prompt, and more accurate function calling.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000006","input_cache_read":"0.0000002"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2024-07-31","expiration_date":null},{"id":"mistralai/pixtral-large-2411","canonical_slug":"mistralai/pixtral-large-2411","hugging_face_id":"","name":"Mistral: Pixtral Large 2411","created":1731977388,"description":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411). The model is able to understand documents, charts and natural images.\n\nThe model is available under the Mistral Research License (MRL) for research and educational use, and the Mistral Commercial License for experimentation, testing, and production for commercial purposes.\n\n","context_length":128000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000006","input_cache_read":"0.0000002"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2024-07-31","expiration_date":null},{"id":"x-ai/grok-4.20-multi-agent","canonical_slug":"x-ai/grok-4.20-multi-agent-20260309","hugging_face_id":"","name":"xAI: Grok 4.20 Multi-Agent","created":1774979158,"description":"Grok 4.20 Multi-Agent is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows. Multiple agents operate in parallel to conduct deep research, coordinate tool use, and synthesize information across complex tasks.\n\nReasoning effort behavior:\n- low / medium: 4 agents\n- high / xhigh: 16 agents","context_length":2000000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000006","web_search":"0.005","input_cache_read":"0.0000002"},"top_provider":{"context_length":2000000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","logprobs","max_tokens","reasoning","response_format","seed","structured_outputs","temperature","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-09-01","expiration_date":null},{"id":"x-ai/grok-4.20","canonical_slug":"x-ai/grok-4.20-20260309","hugging_face_id":"","name":"xAI: Grok 4.20","created":1774979019,"description":"Grok 4.20 is xAI's newest flagship model 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.\n\nReasoning can be enabled/disabled using the `reasoning` `enabled` parameter in the API. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#controlling-reasoning-tokens)","context_length":2000000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000006","web_search":"0.005","input_cache_read":"0.0000002"},"top_provider":{"context_length":2000000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","logprobs","max_tokens","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-09-01","expiration_date":null},{"id":"mistralai/mixtral-8x22b-instruct","canonical_slug":"mistralai/mixtral-8x22b-instruct","hugging_face_id":"mistralai/Mixtral-8x22B-Instruct-v0.1","name":"Mistral: Mixtral 8x22B Instruct","created":1713312000,"description":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include:\n- strong math, coding, and reasoning\n- large context length (64k)\n- fluency in English, French, Italian, German, and Spanish\n\nSee benchmarks on the launch announcement [here](https://mistral.ai/news/mixtral-8x22b/).\n#moe","context_length":65536,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":"mistral"},"pricing":{"prompt":"0.000002","completion":"0.000006","input_cache_read":"0.0000002"},"top_provider":{"context_length":65536,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2024-01-31","expiration_date":null},{"id":"mistralai/mistral-large","canonical_slug":"mistralai/mistral-large","hugging_face_id":null,"name":"Mistral Large","created":1708905600,"description":"This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/).\n\nIt supports dozens of languages including French, German, Spanish, Italian, Portuguese, Arabic, Hindi, Russian, Chinese, Japanese, and Korean, along with 80+ coding languages including Python, Java, C, C++, JavaScript, and Bash. Its long context window allows precise information recall from large documents.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000006","input_cache_read":"0.0000002"},"top_provider":{"context_length":128000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2024-11-30","expiration_date":null},{"id":"mistralai/mistral-large-2407","canonical_slug":"mistralai/mistral-large-2407","hugging_face_id":"","name":"Mistral Large 2407","created":1731978415,"description":"This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/).\n\nIt supports dozens of languages including French, German, Spanish, Italian, Portuguese, Arabic, Hindi, Russian, Chinese, Japanese, and Korean, along with 80+ coding languages including Python, Java, C, C++, JavaScript, and Bash. Its long context window allows precise information recall from large documents.\n","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Mistral","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000006","input_cache_read":"0.0000002"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":0.3},"knowledge_cutoff":"2024-03-31","expiration_date":null},{"id":"openai/gpt-4.1","canonical_slug":"openai/gpt-4.1-2025-04-14","hugging_face_id":"","name":"OpenAI: GPT-4.1","created":1744651385,"description":"GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning. It supports a 1 million token context window and outperforms GPT-4o and GPT-4.5 across coding (54.6% SWE-bench Verified), instruction compliance (87.4% IFEval), and multimodal understanding benchmarks. It is tuned for precise code diffs, agent reliability, and high recall in large document contexts, making it ideal for agents, IDE tooling, and enterprise knowledge retrieval.","context_length":1047576,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000008","web_search":"0.01","input_cache_read":"0.0000005"},"top_provider":{"context_length":1047576,"max_completion_tokens":32768,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"openai/o3","canonical_slug":"openai/o3-2025-04-16","hugging_face_id":"","name":"OpenAI: o3","created":1744823457,"description":"o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following. Use it to think through multi-step problems that involve analysis across text, code, and images. ","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000008","web_search":"0.01","input_cache_read":"0.0000005"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"openai/o4-mini-deep-research","canonical_slug":"openai/o4-mini-deep-research-2025-06-26","hugging_face_id":"","name":"OpenAI: o4 Mini Deep Research","created":1760129642,"description":"o4-mini-deep-research is OpenAI's faster, more affordable deep research model—ideal for tackling complex, multi-step research tasks.\n\nNote: This model always uses the 'web_search' tool which adds additional cost.","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000008","web_search":"0.01","input_cache_read":"0.0000005"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","presence_penalty","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-3.1-pro-preview","canonical_slug":"google/gemini-3.1-pro-preview-20260219","hugging_face_id":"","name":"Google: Gemini 3.1 Pro Preview","created":1771509627,"description":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation of the Gemini 3 series, it combines high-precision reasoning across text, image, video, audio, and code with a 1M-token context window. Reasoning Details must be preserved when using multi-turn tool calling, see our docs here: https://openrouter.ai/docs/use-cases/reasoning-tokens#preserving-reasoning. The 3.1 update introduces measurable gains in SWE benchmarks and real-world coding environments, along with stronger autonomous task execution in structured domains such as finance and spreadsheet-based workflows.\n\nDesigned for advanced development and agentic systems, Gemini 3.1 Pro Preview improves long-horizon stability and tool orchestration while increasing token efficiency. It introduces a new medium thinking level to better balance cost, speed, and performance. The model excels in agentic coding, structured planning, multimodal analysis, and workflow automation, making it well-suited for autonomous agents, financial modeling, spreadsheet automation, and high-context enterprise tasks.","context_length":1048576,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["audio","file","image","text","video"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000012","image":"0.000002","audio":"0.000002","internal_reasoning":"0.000012","input_cache_read":"0.0000002","input_cache_write":"0.000000375"},"top_provider":{"context_length":1048576,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-3-pro-image-preview","canonical_slug":"google/gemini-3-pro-image-preview-20251120","hugging_face_id":"","name":"Google: Nano Banana Pro (Gemini 3 Pro Image Preview)","created":1763653797,"description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro. It extends the original Nano Banana with significantly improved multimodal reasoning, real-world grounding, and high-fidelity visual synthesis. The model generates context-rich graphics, from infographics and diagrams to cinematic composites, and can incorporate real-time information via Search grounding.\n\nIt offers industry-leading text rendering in images (including long passages and multilingual layouts), consistent multi-image blending, and accurate identity preservation across up to five subjects. Nano Banana Pro adds fine-grained creative controls such as localized edits, lighting and focus adjustments, camera transformations, and support for 2K/4K outputs and flexible aspect ratios. It is designed for professional-grade design, product visualization, storyboarding, and complex multi-element compositions while remaining efficient for general image creation workflows.","context_length":65536,"architecture":{"modality":"text+image->text+image","input_modalities":["image","text"],"output_modalities":["image","text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000012","image":"0.000002","audio":"0.000002","internal_reasoning":"0.000012","input_cache_read":"0.0000002","input_cache_write":"0.000000375"},"top_provider":{"context_length":65536,"max_completion_tokens":32768,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"google/gemini-3.1-pro-preview-customtools","canonical_slug":"google/gemini-3.1-pro-preview-customtools-20260219","hugging_face_id":null,"name":"Google: Gemini 3.1 Pro Preview Custom Tools","created":1772045923,"description":"Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool when more efficient third-party or user-defined functions are available. This specialized preview endpoint significantly increases function calling reliability and ensures the model selects the most appropriate tool in coding agents and complex, multi-tool workflows.\n\nIt retains the core strengths of Gemini 3.1 Pro, including multimodal reasoning across text, image, video, audio, and code, a 1M-token context window, and strong software engineering performance.","context_length":1048756,"architecture":{"modality":"text+image+file+audio+video->text","input_modalities":["text","audio","image","video","file"],"output_modalities":["text"],"tokenizer":"Gemini","instruct_type":null},"pricing":{"prompt":"0.000002","completion":"0.000012","image":"0.000002","audio":"0.000002","internal_reasoning":"0.000012","input_cache_read":"0.0000002","input_cache_write":"0.000000375"},"top_provider":{"context_length":1048576,"max_completion_tokens":65536,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-5-image-mini","canonical_slug":"openai/gpt-5-image-mini","hugging_face_id":"","name":"OpenAI: GPT-5 Image Mini","created":1760624583,"description":"GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation. This natively multimodal model features superior instruction following, text rendering, and detailed image editing with reduced latency and cost. It excels at high-quality visual creation while maintaining strong text understanding, making it ideal for applications that require both efficient image generation and text processing at scale.","context_length":400000,"architecture":{"modality":"text+image+file->text+image","input_modalities":["file","image","text"],"output_modalities":["image","text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.000002","web_search":"0.01","input_cache_read":"0.00000025"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","presence_penalty","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-4o-search-preview","canonical_slug":"openai/gpt-4o-search-preview-2025-03-11","hugging_face_id":"","name":"OpenAI: GPT-4o Search Preview","created":1741817949,"description":"GPT-4o Search Previewis a specialized model for web search in Chat Completions. It is trained to understand and execute web search queries.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001","web_search":"0.035"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","structured_outputs","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/gpt-4o-audio-preview","canonical_slug":"openai/gpt-4o-audio-preview","hugging_face_id":"","name":"OpenAI: GPT-4o Audio","created":1755233061,"description":"The gpt-4o-audio-preview model adds support for audio inputs as prompts. This enhancement allows the model to detect nuances within audio recordings and add depth to generated user experiences. Audio outputs are currently not supported. Audio tokens are priced at $40 per million input and $80 per million output audio tokens.","context_length":128000,"architecture":{"modality":"text+audio->text+audio","input_modalities":["audio","text"],"output_modalities":["text","audio"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001","audio":"0.00004"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/gpt-audio","canonical_slug":"openai/gpt-audio","hugging_face_id":"","name":"OpenAI: GPT Audio","created":1768862569,"description":"The gpt-audio model is OpenAI's first generally available audio model. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Audio is priced at $32 per million input tokens and $64 per million output tokens.","context_length":128000,"architecture":{"modality":"text+audio->text+audio","input_modalities":["text","audio"],"output_modalities":["text","audio"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001","audio":"0.000032"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"cohere/command-a","canonical_slug":"cohere/command-a-03-2025","hugging_face_id":"CohereForAI/c4ai-command-a-03-2025","name":"Cohere: Command A","created":1741894342,"description":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases.\nCompared to other leading proprietary and open-weights models Command A delivers maximum performance with minimum hardware costs, excelling on business-critical agentic and multilingual tasks.","context_length":256000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001"},"top_provider":{"context_length":256000,"max_completion_tokens":8192,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-08-31","expiration_date":null},{"id":"openai/gpt-4o-2024-11-20","canonical_slug":"openai/gpt-4o-2024-11-20","hugging_face_id":"","name":"OpenAI: GPT-4o (2024-11-20)","created":1732127594,"description":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability. It’s also better at working with uploaded files, providing deeper insights & more thorough responses.\n\nGPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001","input_cache_read":"0.00000125"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/gpt-4o-2024-08-06","canonical_slug":"openai/gpt-4o-2024-08-06","hugging_face_id":null,"name":"OpenAI: GPT-4o (2024-08-06)","created":1722902400,"description":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format. Read more [here](https://openai.com/index/introducing-structured-outputs-in-the-api/).\n\nGPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001","input_cache_read":"0.00000125"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/gpt-4o","canonical_slug":"openai/gpt-4o","hugging_face_id":null,"name":"OpenAI: GPT-4o","created":1715558400,"description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)\n\n#multimodal","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001","input_cache_read":"0.00000125"},"top_provider":{"context_length":128000,"max_completion_tokens":16384,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"cohere/command-r-plus-08-2024","canonical_slug":"cohere/command-r-plus-08-2024","hugging_face_id":null,"name":"Cohere: Command R+ (08-2024)","created":1724976000,"description":"command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint the same.\n\nRead the launch post [here](https://docs.cohere.com/changelog/command-gets-refreshed).\n\nUse of this model is subject to Cohere's [Usage Policy](https://docs.cohere.com/docs/usage-policy) and [SaaS Agreement](https://cohere.com/saas-agreement).","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Cohere","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.00001"},"top_provider":{"context_length":128000,"max_completion_tokens":4000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{},"knowledge_cutoff":"2024-03-31","expiration_date":null},{"id":"openai/gpt-5.4","canonical_slug":"openai/gpt-5.4-20260305","hugging_face_id":"","name":"OpenAI: GPT-5.4","created":1772734352,"description":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system. It features a 1M+ token context window (922K input, 128K output) with support for text and image inputs, enabling high-context reasoning, coding, and multimodal analysis within the same workflow.\n\nThe model delivers improved performance in coding, document understanding, tool use, and instruction following. It is designed as a strong default for both general-purpose tasks and software engineering, capable of generating production-quality code, synthesizing information across multiple sources, and executing complex multi-step workflows with fewer iterations and greater token efficiency.","context_length":1050000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.0000025","completion":"0.000015","web_search":"0.01","input_cache_read":"0.00000025"},"top_provider":{"context_length":1050000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-3.5-turbo-16k","canonical_slug":"openai/gpt-3.5-turbo-16k","hugging_face_id":null,"name":"OpenAI: GPT-3.5 Turbo 16k","created":1693180800,"description":"This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost. Training data: up to Sep 2021.","context_length":16385,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000003","completion":"0.000004"},"top_provider":{"context_length":16385,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2021-09-30","expiration_date":null},{"id":"anthropic/claude-sonnet-4.6","canonical_slug":"anthropic/claude-4.6-sonnet-20260217","hugging_face_id":"","name":"Anthropic: Claude Sonnet 4.6","created":1771342990,"description":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with memory, polished document creation, and confident computer use for web QA and workflow automation.","context_length":1000000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Claude","instruct_type":null},"pricing":{"prompt":"0.000003","completion":"0.000015","web_search":"0.01","input_cache_read":"0.0000003","input_cache_write":"0.00000375"},"top_provider":{"context_length":1000000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p","verbosity"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"x-ai/grok-3","canonical_slug":"x-ai/grok-3","hugging_face_id":"","name":"xAI: Grok 3","created":1749582908,"description":"Grok 3 is the latest model from xAI. It's their flagship model that excels at enterprise use cases like data extraction, coding, and text summarization. Possesses deep domain knowledge in finance, healthcare, law, and science.\n\n","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.000003","completion":"0.000015","web_search":"0.005","input_cache_read":"0.00000075"},"top_provider":{"context_length":131072,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2025-02-28","expiration_date":null},{"id":"perplexity/sonar-pro-search","canonical_slug":"perplexity/sonar-pro-search","hugging_face_id":"","name":"Perplexity: Sonar Pro Search","created":1761854366,"description":"Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system. It is designed for deeper reasoning and analysis. Pricing is based on tokens plus $18 per thousand requests. This model powers the Pro Search mode on the Perplexity platform.\n\nSonar Pro Search adds autonomous, multi-step reasoning to Sonar Pro. So, instead of just one query + synthesis, it plans and executes entire research workflows using tools.","context_length":200000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Other","instruct_type":null},"pricing":{"prompt":"0.000003","completion":"0.000015","web_search":"0.018"},"top_provider":{"context_length":200000,"max_completion_tokens":8000,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","max_tokens","presence_penalty","reasoning","structured_outputs","temperature","top_k","top_p","web_search_options"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"anthropic/claude-sonnet-4.5","canonical_slug":"anthropic/claude-4.5-sonnet-20250929","hugging_face_id":"","name":"Anthropic: Claude Sonnet 4.5","created":1759161676,"description":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with improvements across system design, code security, and specification adherence. The model is designed for extended autonomous operation, maintaining task continuity across sessions and providing fact-based progress tracking.\n\nSonnet 4.5 also introduces stronger agentic capabilities, including improved tool orchestration, speculative parallel execution, and more efficient context and memory management. With enhanced context tracking and awareness of token usage across tool calls, it is particularly well-suited for multi-context and long-running workflows. Use cases span software engineering, cybersecurity, financial analysis, research agents, and other domains requiring sustained reasoning and tool use.","context_length":1000000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"Claude","instruct_type":null},"pricing":{"prompt":"0.000003","completion":"0.000015","web_search":"0.01","input_cache_read":"0.0000003","input_cache_write":"0.00000375"},"top_provider":{"context_length":1000000,"max_completion_tokens":64000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":1,"top_p":1,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null},{"id":"x-ai/grok-4","canonical_slug":"x-ai/grok-4-07-09","hugging_face_id":"","name":"xAI: Grok 4","created":1752087689,"description":"Grok 4 is xAI's latest reasoning model with a 256k context window. It supports parallel tool calling, structured outputs, and both image and text inputs. Note that reasoning is not exposed, reasoning cannot be disabled, and the reasoning effort cannot be specified. Pricing increases once the total tokens in a given request is greater than 128k tokens. See more details on the [xAI docs](https://docs.x.ai/docs/models/grok-4-0709)","context_length":256000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"Grok","instruct_type":null},"pricing":{"prompt":"0.000003","completion":"0.000015","web_search":"0.005","input_cache_read":"0.00000075"},"top_provider":{"context_length":256000,"max_completion_tokens":null,"is_moderated":false},"per_request_limits":null,"supported_parameters":["include_reasoning","logprobs","max_tokens","reasoning","response_format","seed","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2025-07-31","expiration_date":null},{"id":"openai/gpt-4o-2024-05-13","canonical_slug":"openai/gpt-4o-2024-05-13","hugging_face_id":null,"name":"OpenAI: GPT-4o (2024-05-13)","created":1715558400,"description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)\n\n#multimodal","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000005","completion":"0.000015"},"top_provider":{"context_length":128000,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"anthropic/claude-opus-4.5","canonical_slug":"anthropic/claude-4.5-opus-20251124","hugging_face_id":"","name":"Anthropic: Claude Opus 4.5","created":1764010580,"description":"Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use. It offers strong multimodal capabilities, competitive performance across real-world coding and reasoning benchmarks, and improved robustness to prompt injection. The model is designed to operate efficiently across varied effort levels, enabling developers to trade off speed, depth, and token usage depending on task requirements. It comes with a new parameter to control token efficiency, which can be accessed using the OpenRouter Verbosity parameter with low, medium, or high.\n\nOpus 4.5 supports advanced tool use, extended context management, and coordinated multi-agent setups, making it well-suited for autonomous research, debugging, multi-step planning, and spreadsheet/browser manipulation. It delivers substantial gains in structured reasoning, execution reliability, and alignment compared to prior Opus generations, while reducing token overhead and improving performance on long-running tasks.","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"Claude","instruct_type":null},"pricing":{"prompt":"0.000005","completion":"0.000025","web_search":"0.01","input_cache_read":"0.0000005","input_cache_write":"0.00000625"},"top_provider":{"context_length":200000,"max_completion_tokens":64000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","verbosity"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"anthropic/claude-opus-4.6","canonical_slug":"anthropic/claude-4.6-opus-20260205","hugging_face_id":"","name":"Anthropic: Claude Opus 4.6","created":1770219050,"description":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective for large codebases, complex refactors, and multi-step debugging that unfolds over time. The model shows deeper contextual understanding, stronger problem decomposition, and greater reliability on hard engineering tasks than prior generations.\n\nBeyond coding, Opus 4.6 excels at sustained knowledge work. It produces near-production-ready documents, plans, and analyses in a single pass, and maintains coherence across very long outputs and extended sessions. This makes it a strong default for tasks that require persistence, judgment, and follow-through, such as technical design, migration planning, and end-to-end project execution.\n\nFor users upgrading from earlier Opus versions, see our [official migration guide here](https://openrouter.ai/docs/guides/guides/model-migrations/claude-4-6-opus)\n","context_length":1000000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"Claude","instruct_type":null},"pricing":{"prompt":"0.000005","completion":"0.000025","web_search":"0.01","input_cache_read":"0.0000005","input_cache_write":"0.00000625"},"top_provider":{"context_length":1000000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p","verbosity"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-4o:extended","canonical_slug":"openai/gpt-4o","hugging_face_id":null,"name":"OpenAI: GPT-4o (extended)","created":1715558400,"description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.\n\nFor benchmarking against other models, it was briefly called [\"im-also-a-good-gpt2-chatbot\"](https://twitter.com/LiamFedus/status/1790064963966370209)\n\n#multimodal","context_length":128000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000006","completion":"0.000018"},"top_provider":{"context_length":128000,"max_completion_tokens":64000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p","web_search_options"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/gpt-5-image","canonical_slug":"openai/gpt-5-image","hugging_face_id":"","name":"OpenAI: GPT-5 Image","created":1760447986,"description":"[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities. It offers major improvements in reasoning, code quality, and user experience while incorporating GPT Image 1's superior instruction following, text rendering, and detailed image editing.","context_length":400000,"architecture":{"modality":"text+image+file->text+image","input_modalities":["image","text","file"],"output_modalities":["image","text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00001","completion":"0.00001","web_search":"0.01","input_cache_read":"0.00000125"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","presence_penalty","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-4-1106-preview","canonical_slug":"openai/gpt-4-1106-preview","hugging_face_id":null,"name":"OpenAI: GPT-4 Turbo (older v1106)","created":1699228800,"description":"The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling.\n\nTraining data: up to April 2023.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00001","completion":"0.00003"},"top_provider":{"context_length":128000,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-04-30","expiration_date":null},{"id":"openai/gpt-4-turbo","canonical_slug":"openai/gpt-4-turbo","hugging_face_id":null,"name":"OpenAI: GPT-4 Turbo","created":1712620800,"description":"The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling.\n\nTraining data: up to December 2023.","context_length":128000,"architecture":{"modality":"text+image->text","input_modalities":["text","image"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00001","completion":"0.00003"},"top_provider":{"context_length":128000,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"openai/gpt-4-turbo-preview","canonical_slug":"openai/gpt-4-turbo-preview","hugging_face_id":null,"name":"OpenAI: GPT-4 Turbo Preview","created":1706140800,"description":"The preview GPT-4 model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more. Training data: up to Dec 2023.\n\n**Note:** heavily rate limited by OpenAI while in preview.","context_length":128000,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00001","completion":"0.00003"},"top_provider":{"context_length":128000,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2023-12-31","expiration_date":null},{"id":"openai/o3-deep-research","canonical_slug":"openai/o3-deep-research-2025-06-26","hugging_face_id":"","name":"OpenAI: o3 Deep Research","created":1760129661,"description":"o3-deep-research is OpenAI's advanced model for deep research, designed to tackle complex, multi-step research tasks.\n\nNote: This model always uses the 'web_search' tool which adds additional cost.","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00001","completion":"0.00004","web_search":"0.01","input_cache_read":"0.0000025"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","logprobs","max_tokens","presence_penalty","reasoning","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/o1","canonical_slug":"openai/o1-2024-12-17","hugging_face_id":"","name":"OpenAI: o1","created":1734459999,"description":"The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding. The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought. \n\nThe o1 models are optimized for math, science, programming, and other STEM-related tasks. They consistently exhibit PhD-level accuracy on benchmarks in physics, chemistry, and biology. Learn more in the [launch announcement](https://openai.com/o1).\n","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000015","completion":"0.00006","input_cache_read":"0.0000075"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"openai/gpt-5-pro","canonical_slug":"openai/gpt-5-pro-2025-10-06","hugging_face_id":"","name":"OpenAI: GPT-5 Pro","created":1759776663,"description":"GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. It supports test-time routing features and advanced prompt understanding, including user-specified intent like \"think hard about this.\" Improvements include reductions in hallucination, sycophancy, and better performance in coding, writing, and health-related tasks.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000015","completion":"0.00012","web_search":"0.01"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-09-30","expiration_date":null},{"id":"openai/o3-pro","canonical_slug":"openai/o3-pro-2025-06-10","hugging_face_id":"","name":"OpenAI: o3 Pro","created":1749598352,"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.\n\nNote that BYOK is required for this model. Set up here: https://openrouter.ai/settings/integrations","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","file","image"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00002","completion":"0.00008","web_search":"0.01"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2024-06-30","expiration_date":null},{"id":"openai/gpt-5.2-pro","canonical_slug":"openai/gpt-5.2-pro-20251211","hugging_face_id":"","name":"OpenAI: GPT-5.2 Pro","created":1765389780,"description":"GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. It supports test-time routing features and advanced prompt understanding, including user-specified intent like \"think hard about this.\" Improvements include reductions in hallucination, sycophancy, and better performance in coding, writing, and health-related tasks.","context_length":400000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.000021","completion":"0.000168","web_search":"0.01"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/gpt-4","canonical_slug":"openai/gpt-4","hugging_face_id":null,"name":"OpenAI: GPT-4","created":1685232000,"description":"OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and advanced reasoning capabilities. Training data: up to Sep 2021.","context_length":8191,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00003","completion":"0.00006"},"top_provider":{"context_length":8191,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2021-09-30","expiration_date":null},{"id":"openai/gpt-4-0314","canonical_slug":"openai/gpt-4-0314","hugging_face_id":null,"name":"OpenAI: GPT-4 (older v0314)","created":1685232000,"description":"GPT-4-0314 is the first version of GPT-4 released, with a context length of 8,192 tokens, and was supported until June 14. Training data: up to Sep 2021.","context_length":8191,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00003","completion":"0.00006"},"top_provider":{"context_length":8191,"max_completion_tokens":4096,"is_moderated":true},"per_request_limits":null,"supported_parameters":["frequency_penalty","logit_bias","logprobs","max_tokens","presence_penalty","response_format","seed","stop","structured_outputs","temperature","tool_choice","tools","top_logprobs","top_p"],"default_parameters":{},"knowledge_cutoff":"2021-09-30","expiration_date":null},{"id":"openai/gpt-5.4-pro","canonical_slug":"openai/gpt-5.4-pro-20260305","hugging_face_id":"","name":"OpenAI: GPT-5.4 Pro","created":1772734366,"description":"GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks. It features a 1M+ token context window (922K input, 128K output) with support for text and image inputs. Optimized for step-by-step reasoning, instruction following, and accuracy, GPT-5.4 Pro excels at agentic coding, long-context workflows, and multi-step problem solving.","context_length":1050000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00003","completion":"0.00018","web_search":"0.01"},"top_provider":{"context_length":1050000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs","tool_choice","tools"],"default_parameters":{"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"openai/o1-pro","canonical_slug":"openai/o1-pro","hugging_face_id":"","name":"OpenAI: o1-pro","created":1742423211,"description":"The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o1-pro model uses more compute to think harder and provide consistently better answers.","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["text","image","file"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00015","completion":"0.0006"},"top_provider":{"context_length":200000,"max_completion_tokens":100000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","seed","structured_outputs"],"default_parameters":{},"knowledge_cutoff":"2023-10-31","expiration_date":null},{"id":"deepseek/deepseek-v3.2-speciale","canonical_slug":"deepseek/deepseek-v3.2-speciale-20251201","hugging_face_id":"deepseek-ai/DeepSeek-V3.2-Speciale","name":"DeepSeek: DeepSeek V3.2 Speciale","created":1764594837,"description":"DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance. It builds on DeepSeek Sparse Attention (DSA) for efficient long-context processing, then scales post-training reinforcement learning to push capability beyond the base model. Reported evaluations place Speciale ahead of GPT-5 on difficult reasoning workloads, with proficiency comparable to Gemini-3.0-Pro, while retaining strong coding and tool-use reliability. Like V3.2, it benefits from a large-scale agentic task synthesis pipeline that improves compliance and generalization in interactive environments.","context_length":131072,"architecture":{"modality":"text->text","input_modalities":["text"],"output_modalities":["text"],"tokenizer":"DeepSeek","instruct_type":null},"pricing":{"prompt":"0.0000004","completion":"0.0000012","input_cache_read":"0.0000002"},"top_provider":{"context_length":163840,"max_completion_tokens":163840,"is_moderated":false},"per_request_limits":null,"supported_parameters":["frequency_penalty","include_reasoning","logit_bias","max_tokens","min_p","presence_penalty","reasoning","repetition_penalty","response_format","seed","stop","structured_outputs","temperature","top_k","top_p"],"default_parameters":{"temperature":1,"top_p":0.95,"frequency_penalty":null},"knowledge_cutoff":null,"expiration_date":null},{"id":"anthropic/claude-opus-4.1","canonical_slug":"anthropic/claude-4.1-opus-20250805","hugging_face_id":"","name":"Anthropic: Claude Opus 4.1","created":1754411591,"description":"Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks. It achieves 74.5% on SWE-bench Verified and shows notable gains in multi-file code refactoring, debugging precision, and detail-oriented reasoning. The model supports extended thinking up to 64K tokens and is optimized for tasks involving research, data analysis, and tool-assisted reasoning.","context_length":200000,"architecture":{"modality":"text+image+file->text","input_modalities":["image","text","file"],"output_modalities":["text"],"tokenizer":"Claude","instruct_type":null},"pricing":{"prompt":"0.000015","completion":"0.000075","web_search":"0.01","input_cache_read":"0.0000015","input_cache_write":"0.00001875"},"top_provider":{"context_length":200000,"max_completion_tokens":32000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["include_reasoning","max_tokens","reasoning","response_format","stop","structured_outputs","temperature","tool_choice","tools","top_k","top_p"],"default_parameters":{"temperature":null,"top_p":null,"frequency_penalty":null},"knowledge_cutoff":"2025-01-31","expiration_date":null}]}