{"success":true,"data":[{"model_name":"sora-2","description":"OpenAI 视频生成标准模型，按生成秒数计费，适合 720p 视频生成和批处理视频任务。","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"openai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0.1,"context_length":null}},{"model_name":"Veo 3.1","description":"Google Gemini 视频生成标准模型，按生成秒数、速度档和分辨率计费。","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"视频生成","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,视频生成","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0.4,"context_length":null}},{"model_name":"~deepseek/deepseek-flash-latest","description":"This model always redirects to the latest model in the DeepSeek Flash family.","quota_type":0,"model_ratio":0.003,"model_price":0,"owner_by":"~deepseek","completion_ratio":83.3333,"cache_ratio":0.833333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~deepseek","grokcode":{"input_usd_per_m":0.006,"output_usd_per_m":0.5,"openrouter_id":"~deepseek/deepseek-flash-latest","context_length":1048576}},{"model_name":"deepseek-flash-latest","description":"This model always redirects to the latest model in the DeepSeek Flash family.","quota_type":0,"model_ratio":0.003,"model_price":0,"owner_by":"~deepseek","completion_ratio":83.3333,"cache_ratio":0.833333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~deepseek","grokcode":{"input_usd_per_m":0.006,"output_usd_per_m":0.5,"openrouter_id":"~deepseek/deepseek-flash-latest","context_length":1048576}},{"model_name":"glm-5.2","description":"GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...","quota_type":0,"model_ratio":0.003,"model_price":0,"owner_by":"glm","completion_ratio":1133.33,"cache_ratio":0.833333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.006,"output_usd_per_m":6.8,"openrouter_id":"z-ai/glm-5.2","context_length":1048576}},{"model_name":"z-ai/glm-5.2","description":"GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...","quota_type":0,"model_ratio":0.003,"model_price":0,"owner_by":"glm","completion_ratio":1133.33,"cache_ratio":0.833333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.006,"output_usd_per_m":6.8,"openrouter_id":"z-ai/glm-5.2","context_length":1048576}},{"model_name":"inclusionai/ling-3.0-tiny","description":"Ling 3.0 Tiny is a mixture-of-experts model from InclusionAI, with 1.3B active parameters out of 7.9B total. It is designed for responsive agents, instruction following, and multi-turn conversations, with switchable...","quota_type":0,"model_ratio":0.0075,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.015,"output_usd_per_m":0.045,"openrouter_id":"inclusionai/ling-3.0-tiny","context_length":262144}},{"model_name":"ling-3.0-tiny","description":"Ling 3.0 Tiny is a mixture-of-experts model from InclusionAI, with 1.3B active parameters out of 7.9B total. It is designed for responsive agents, instruction following, and multi-turn conversations, with switchable...","quota_type":0,"model_ratio":0.0075,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.015,"output_usd_per_m":0.045,"openrouter_id":"inclusionai/ling-3.0-tiny","context_length":262144}},{"model_name":"inclusionai/ling-3.0-flash","description":"*Ling-3.0-flash* is a *124B-parameter Mixture-of-Experts (MoE) model*, with approximately *5.1B parameters activated per token*. The model is designed with *token efficiency and production-scale agentic inference* as key priorities, enabling developers...","quota_type":0,"model_ratio":0.0105,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.020999999999999998,"output_usd_per_m":0.063,"openrouter_id":"inclusionai/ling-3.0-flash","context_length":262144}},{"model_name":"ling-3.0-flash","description":"*Ling-3.0-flash* is a *124B-parameter Mixture-of-Experts (MoE) model*, with approximately *5.1B parameters activated per token*. The model is designed with *token efficiency and production-scale agentic inference* as key priorities, enabling developers...","quota_type":0,"model_ratio":0.0105,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.020999999999999998,"output_usd_per_m":0.063,"openrouter_id":"inclusionai/ling-3.0-flash","context_length":262144}},{"model_name":"openai/gpt-oss-20b:batch","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...","quota_type":0,"model_ratio":0.012,"model_price":0,"owner_by":"openai","completion_ratio":4.6667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.024,"output_usd_per_m":0.112,"openrouter_id":"openai/gpt-oss-20b:batch","context_length":131072}},{"model_name":"nex-agi/nex-n2.5-mini","description":"Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...","quota_type":0,"model_ratio":0.0125,"model_price":0,"owner_by":"nex-agi","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nex-agi","grokcode":{"input_usd_per_m":0.024999999999999998,"output_usd_per_m":0.09999999999999999,"openrouter_id":"nex-agi/nex-n2.5-mini","context_length":262144}},{"model_name":"nex-n2.5-mini","description":"Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...","quota_type":0,"model_ratio":0.0125,"model_price":0,"owner_by":"nex-agi","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nex-agi","grokcode":{"input_usd_per_m":0.024999999999999998,"output_usd_per_m":0.09999999999999999,"openrouter_id":"nex-agi/nex-n2.5-mini","context_length":262144}},{"model_name":"openai/gpt-5-nano:batch","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...","quota_type":0,"model_ratio":0.0125,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.024999999999999998,"output_usd_per_m":0.19999999999999998,"openrouter_id":"openai/gpt-5-nano:batch","context_length":400000}},{"model_name":"llama-3.2-1b-instruct","description":"Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis. Its smaller size allows it to operate...","quota_type":0,"model_ratio":0.0135,"model_price":0,"owner_by":"llama","completion_ratio":7.4444,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.027,"output_usd_per_m":0.201,"openrouter_id":"meta-llama/llama-3.2-1b-instruct","context_length":60000}},{"model_name":"meta-llama/llama-3.2-1b-instruct","description":"Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis. Its smaller size allows it to operate...","quota_type":0,"model_ratio":0.0135,"model_price":0,"owner_by":"llama","completion_ratio":7.4444,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.027,"output_usd_per_m":0.201,"openrouter_id":"meta-llama/llama-3.2-1b-instruct","context_length":60000}},{"model_name":"~z-ai/glm-flash-latest","description":"This model always redirects to the latest model in the GLM Flash family.","quota_type":0,"model_ratio":0.01425,"model_price":0,"owner_by":"~z-ai","completion_ratio":3.2386,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~z-ai","grokcode":{"input_usd_per_m":0.0285,"output_usd_per_m":0.0923,"openrouter_id":"~z-ai/glm-flash-latest","context_length":1048576}},{"model_name":"glm-flash-latest","description":"This model always redirects to the latest model in the GLM Flash family.","quota_type":0,"model_ratio":0.01425,"model_price":0,"owner_by":"~z-ai","completion_ratio":3.2386,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~z-ai","grokcode":{"input_usd_per_m":0.0285,"output_usd_per_m":0.0923,"openrouter_id":"~z-ai/glm-flash-latest","context_length":1048576}},{"model_name":"mistral-nemo","description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese,...","quota_type":0,"model_ratio":0.0145,"model_price":0,"owner_by":"mistral","completion_ratio":1.0345,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.029,"output_usd_per_m":0.03,"openrouter_id":"mistralai/mistral-nemo","context_length":131072}},{"model_name":"mistralai/mistral-nemo","description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese,...","quota_type":0,"model_ratio":0.0145,"model_price":0,"owner_by":"mistral","completion_ratio":1.0345,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.029,"output_usd_per_m":0.03,"openrouter_id":"mistralai/mistral-nemo","context_length":131072}},{"model_name":"openai/gpt-oss-120b:batch","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...","quota_type":0,"model_ratio":0.0148,"model_price":0,"owner_by":"openai","completion_ratio":4.5946,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.0296,"output_usd_per_m":0.136,"openrouter_id":"openai/gpt-oss-120b:batch","context_length":131072}},{"model_name":"gpt-oss-20b","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...","quota_type":0,"model_ratio":0.015,"model_price":0,"owner_by":"openai","completion_ratio":4.3333,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.03,"output_usd_per_m":0.13,"openrouter_id":"openai/gpt-oss-20b","context_length":131072}},{"model_name":"inference-net/schematron-v2-turbo","description":"Schematron V2 Turbo is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes throughput for high-volume extraction workloads. Extraction instructions must be supplied through a JSON schema in response_format rather...","quota_type":0,"model_ratio":0.015,"model_price":0,"owner_by":"inference-net","completion_ratio":5,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inference-net","grokcode":{"input_usd_per_m":0.03,"output_usd_per_m":0.15,"openrouter_id":"inference-net/schematron-v2-turbo","context_length":128000}},{"model_name":"openai/gpt-oss-20b","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...","quota_type":0,"model_ratio":0.015,"model_price":0,"owner_by":"openai","completion_ratio":4.3333,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.03,"output_usd_per_m":0.13,"openrouter_id":"openai/gpt-oss-20b","context_length":131072}},{"model_name":"qwen/qwen3.7-flash","description":"Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer interaction, with strengths in object recognition, spatial understanding, and real-world...","quota_type":0,"model_ratio":0.015,"model_price":0,"owner_by":"qwen","completion_ratio":4.3333,"cache_ratio":0.2,"create_cache_ratio":1.2667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.03,"output_usd_per_m":0.13,"openrouter_id":"qwen/qwen3.7-flash","context_length":1000000}},{"model_name":"qwen3.7-flash","description":"Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer interaction, with strengths in object recognition, spatial understanding, and real-world...","quota_type":0,"model_ratio":0.015,"model_price":0,"owner_by":"qwen","completion_ratio":4.3333,"cache_ratio":0.2,"create_cache_ratio":1.2667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.03,"output_usd_per_m":0.13,"openrouter_id":"qwen/qwen3.7-flash","context_length":1000000}},{"model_name":"schematron-v2-turbo","description":"Schematron V2 Turbo is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes throughput for high-volume extraction workloads. Extraction instructions must be supplied through a JSON schema in response_format rather...","quota_type":0,"model_ratio":0.015,"model_price":0,"owner_by":"inference-net","completion_ratio":5,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inference-net","grokcode":{"input_usd_per_m":0.03,"output_usd_per_m":0.15,"openrouter_id":"inference-net/schematron-v2-turbo","context_length":128000}},{"model_name":"amazon/nova-micro-v1","description":"Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost. With a context length...","quota_type":0,"model_ratio":0.0175,"model_price":0,"owner_by":"amazon","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.035,"output_usd_per_m":0.14,"openrouter_id":"amazon/nova-micro-v1","context_length":128000}},{"model_name":"nova-micro-v1","description":"Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost. With a context length...","quota_type":0,"model_ratio":0.0175,"model_price":0,"owner_by":"amazon","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.035,"output_usd_per_m":0.14,"openrouter_id":"amazon/nova-micro-v1","context_length":128000}},{"model_name":"cohere/command-r7b-12-2024","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...","quota_type":0,"model_ratio":0.01875,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0.0375,"output_usd_per_m":0.15,"openrouter_id":"cohere/command-r7b-12-2024","context_length":128000}},{"model_name":"command-r7b-12-2024","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...","quota_type":0,"model_ratio":0.01875,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0.0375,"output_usd_per_m":0.15,"openrouter_id":"cohere/command-r7b-12-2024","context_length":128000}},{"model_name":"extract","description":"Extract is a description-driven entity extraction model from ScaleDown. Instead of a fixed set of entity types, callers describe the fields they want in plain English through `provider.options.scaledown.entities`, and the...","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"scaledown","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,scaledown","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0,"openrouter_id":"scaledown/extract","context_length":1000000}},{"model_name":"inception/mercury-2.5","description":"Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception. Instead of generating tokens sequentially, Mercury 2.5 produces and refines multiple tokens in parallel, achieving...","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"inception","completion_ratio":3.75,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inception","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0.15,"openrouter_id":"inception/mercury-2.5","context_length":260000}},{"model_name":"l3-lunaris-8b","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....","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"sao10k","completion_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sao10k","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0.049999999999999996,"openrouter_id":"sao10k/l3-lunaris-8b","context_length":8192}},{"model_name":"mercury-2.5","description":"Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception. Instead of generating tokens sequentially, Mercury 2.5 produces and refines multiple tokens in parallel, achieving...","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"inception","completion_ratio":3.75,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inception","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0.15,"openrouter_id":"inception/mercury-2.5","context_length":260000}},{"model_name":"sao10k/l3-lunaris-8b","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....","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"sao10k","completion_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sao10k","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0.049999999999999996,"openrouter_id":"sao10k/l3-lunaris-8b","context_length":8192}},{"model_name":"scaledown/extract","description":"Extract is a description-driven entity extraction model from ScaleDown. Instead of a fixed set of entity types, callers describe the fields they want in plain English through `provider.options.scaledown.entities`, and the...","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"scaledown","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,scaledown","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0,"openrouter_id":"scaledown/extract","context_length":1000000}},{"model_name":"scaledown/summarize","description":"Summarize is an abstractive summarization model from ScaleDown. It condenses long text and documents into a fluent, faithful rewrite rather than lifting sentences from the source, and is suited for...","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"scaledown","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,scaledown","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0,"openrouter_id":"scaledown/summarize","context_length":1000000}},{"model_name":"summarize","description":"Summarize is an abstractive summarization model from ScaleDown. It condenses long text and documents into a fluent, faithful rewrite rather than lifting sentences from the source, and is suited for...","quota_type":0,"model_ratio":0.02,"model_price":0,"owner_by":"scaledown","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,scaledown","grokcode":{"input_usd_per_m":0.04,"output_usd_per_m":0,"openrouter_id":"scaledown/summarize","context_length":1000000}},{"model_name":"~deepseek/deepseek-v4-flash-latest","description":"This model always redirects to the latest model in the DeepSeek V4 Flash family.","quota_type":0,"model_ratio":0.0217,"model_price":0,"owner_by":"~deepseek","completion_ratio":6.9124,"cache_ratio":0.806452,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~deepseek","grokcode":{"input_usd_per_m":0.0434,"output_usd_per_m":0.3,"openrouter_id":"~deepseek/deepseek-v4-flash-latest","context_length":1048576}},{"model_name":"deepseek-v4-flash-0731","description":"DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....","quota_type":0,"model_ratio":0.0217,"model_price":0,"owner_by":"deepseek","completion_ratio":6.9124,"cache_ratio":0.806452,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.0434,"output_usd_per_m":0.3,"openrouter_id":"deepseek/deepseek-v4-flash-0731","context_length":1048576}},{"model_name":"deepseek-v4-flash-latest","description":"This model always redirects to the latest model in the DeepSeek V4 Flash family.","quota_type":0,"model_ratio":0.0217,"model_price":0,"owner_by":"~deepseek","completion_ratio":6.9124,"cache_ratio":0.806452,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~deepseek","grokcode":{"input_usd_per_m":0.0434,"output_usd_per_m":0.3,"openrouter_id":"~deepseek/deepseek-v4-flash-latest","context_length":1048576}},{"model_name":"deepseek/deepseek-v4-flash-0731","description":"DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....","quota_type":0,"model_ratio":0.0217,"model_price":0,"owner_by":"deepseek","completion_ratio":6.9124,"cache_ratio":0.806452,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.0434,"output_usd_per_m":0.3,"openrouter_id":"deepseek/deepseek-v4-flash-0731","context_length":1048576}},{"model_name":"hy-mt2-1.8b","description":"Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided...","quota_type":0,"model_ratio":0.022,"model_price":0,"owner_by":"tencent","completion_ratio":4.0227,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.044,"output_usd_per_m":0.17700000000000002,"openrouter_id":"tencent/hy-mt2-1.8b","context_length":8192}},{"model_name":"tencent/hy-mt2-1.8b","description":"Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided...","quota_type":0,"model_ratio":0.022,"model_price":0,"owner_by":"tencent","completion_ratio":4.0227,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.044,"output_usd_per_m":0.17700000000000002,"openrouter_id":"tencent/hy-mt2-1.8b","context_length":8192}},{"model_name":"deepseek-v4.1-flash","description":"DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...","quota_type":0,"model_ratio":0.0225,"model_price":0,"owner_by":"deepseek","completion_ratio":11.1111,"cache_ratio":0.4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.045,"output_usd_per_m":0.5,"openrouter_id":"deepseek/deepseek-v4.1-flash","context_length":1048576}},{"model_name":"deepseek/deepseek-v4.1-flash","description":"DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...","quota_type":0,"model_ratio":0.0225,"model_price":0,"owner_by":"deepseek","completion_ratio":11.1111,"cache_ratio":0.4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.045,"output_usd_per_m":0.5,"openrouter_id":"deepseek/deepseek-v4.1-flash","context_length":1048576}},{"model_name":"anthropic/claude-haiku-5.5:batch","description":"Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.25,"openrouter_id":"anthropic/claude-haiku-5.5:batch","context_length":1000000}},{"model_name":"gemma-3-12b-it","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,...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"gemma","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.15,"openrouter_id":"google/gemma-3-12b-it","context_length":131072}},{"model_name":"gemma-3-4b-it","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,...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"gemma","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.09999999999999999,"openrouter_id":"google/gemma-3-4b-it","context_length":131072}},{"model_name":"google/gemma-3-12b-it","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,...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"gemma","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.15,"openrouter_id":"google/gemma-3-12b-it","context_length":131072}},{"model_name":"google/gemma-3-4b-it","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,...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"gemma","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.09999999999999999,"openrouter_id":"google/gemma-3-4b-it","context_length":131072}},{"model_name":"gpt-5-nano","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...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.39999999999999997,"openrouter_id":"openai/gpt-5-nano","context_length":400000}},{"model_name":"inference-net/schematron-v2-small","description":"Schematron V2 Small is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes extraction quality for complex schemas and long pages. Extraction instructions must be supplied through a JSON schema...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"inference-net","completion_ratio":4.6,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inference-net","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.22999999999999998,"openrouter_id":"inference-net/schematron-v2-small","context_length":128000}},{"model_name":"lfm-2.5-24b-a2b","description":"LFM2.5-24B-A2B is a mixture-of-experts reasoning model from Liquid AI with 24B total and 2.45B active parameters. It always reasons before answering and is built for agents that call tools. Liquid...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"liquid","completion_ratio":14,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,liquid","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.7,"openrouter_id":"liquid/lfm-2.5-24b-a2b","context_length":128000}},{"model_name":"liquid/lfm-2.5-24b-a2b","description":"LFM2.5-24B-A2B is a mixture-of-experts reasoning model from Liquid AI with 24B total and 2.45B active parameters. It always reasons before answering and is built for agents that call tools. Liquid...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"liquid","completion_ratio":14,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,liquid","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.7,"openrouter_id":"liquid/lfm-2.5-24b-a2b","context_length":128000}},{"model_name":"llama-3.1-8b-instruct","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient. It has demonstrated strong performance compared to...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"llama","completion_ratio":1.6,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.08,"openrouter_id":"meta-llama/llama-3.1-8b-instruct","context_length":131072}},{"model_name":"meta-llama/llama-3.1-8b-instruct","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient. It has demonstrated strong performance compared to...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"llama","completion_ratio":1.6,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.08,"openrouter_id":"meta-llama/llama-3.1-8b-instruct","context_length":131072}},{"model_name":"mistral-small-24b-instruct-2501","description":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"mistral","completion_ratio":1.6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.08,"openrouter_id":"mistralai/mistral-small-24b-instruct-2501","context_length":32768}},{"model_name":"mistralai/mistral-small-24b-instruct-2501","description":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"mistral","completion_ratio":1.6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.08,"openrouter_id":"mistralai/mistral-small-24b-instruct-2501","context_length":32768}},{"model_name":"openai/gpt-4.1-nano:batch","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...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.19999999999999998,"openrouter_id":"openai/gpt-4.1-nano:batch","context_length":1047576}},{"model_name":"openai/gpt-5-nano","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...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.39999999999999997,"openrouter_id":"openai/gpt-5-nano","context_length":400000}},{"model_name":"openai/gpt-6-luna-pro:batch","description":"GPT-6 Luna Pro is the same underlying model as [GPT-6 Luna](https://openrouter.ai/openai/gpt-6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.25,"openrouter_id":"openai/gpt-6-luna-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-6-luna:batch","description":"GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.25,"openrouter_id":"openai/gpt-6-luna:batch","context_length":1050000}},{"model_name":"particle-1","description":"Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as terminal workflows, codebase changes, and data...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"particle-ai","completion_ratio":10,"cache_ratio":0.06,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,particle-ai","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.5,"openrouter_id":"particle-ai/particle-1","context_length":1048576}},{"model_name":"particle-ai/particle-1","description":"Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as terminal workflows, codebase changes, and data...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"particle-ai","completion_ratio":10,"cache_ratio":0.06,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,particle-ai","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.5,"openrouter_id":"particle-ai/particle-1","context_length":1048576}},{"model_name":"schematron-v2-small","description":"Schematron V2 Small is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes extraction quality for complex schemas and long pages. Extraction instructions must be supplied through a JSON schema...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"inference-net","completion_ratio":4.6,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inference-net","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.22999999999999998,"openrouter_id":"inference-net/schematron-v2-small","context_length":128000}},{"model_name":"solar-mini4","description":"Solar Mini 4 is Upstage's compact, cost-efficient language model, a 35B-parameter mixture-of-experts with 3B active parameters and a 524K context window. It is built for agentic use cases where response...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"upstage","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,upstage","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.19999999999999998,"openrouter_id":"upstage/solar-mini4","context_length":524288}},{"model_name":"upstage/solar-mini4","description":"Solar Mini 4 is Upstage's compact, cost-efficient language model, a 35B-parameter mixture-of-experts with 3B active parameters and a 524K context window. It is built for agentic use cases where response...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"upstage","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,upstage","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.19999999999999998,"openrouter_id":"upstage/solar-mini4","context_length":524288}},{"model_name":"z-ai/glm-5.3-flash:batch","description":"GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...","quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"glm","completion_ratio":3.6,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.049999999999999996,"output_usd_per_m":0.18,"openrouter_id":"z-ai/glm-5.3-flash:batch","context_length":1048576}},{"model_name":"llama-3.2-3b-instruct","description":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it...","quota_type":0,"model_ratio":0.02545,"model_price":0,"owner_by":"llama","completion_ratio":6.5815,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.0509,"output_usd_per_m":0.335,"openrouter_id":"meta-llama/llama-3.2-3b-instruct","context_length":131072}},{"model_name":"meta-llama/llama-3.2-3b-instruct","description":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it...","quota_type":0,"model_ratio":0.02545,"model_price":0,"owner_by":"llama","completion_ratio":6.5815,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.0509,"output_usd_per_m":0.335,"openrouter_id":"meta-llama/llama-3.2-3b-instruct","context_length":131072}},{"model_name":"amazon/nova-lite-v1","description":"Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output. Amazon Nova Lite...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"amazon","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.24,"openrouter_id":"amazon/nova-lite-v1","context_length":300000}},{"model_name":"granite-4.2-8b","description":"Granite 4.2 8B is a dense reasoning model from IBM. It is suited for mathematics, code generation, multilingual dialogue, and agentic workflows that need multi-step reasoning. It supports full, low-effort,...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"ibm-granite","completion_ratio":4.1667,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,ibm-granite","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.25,"openrouter_id":"ibm-granite/granite-4.2-8b","context_length":131072}},{"model_name":"ibm-granite/granite-4.2-8b","description":"Granite 4.2 8B is a dense reasoning model from IBM. It is suited for mathematics, code generation, multilingual dialogue, and agentic workflows that need multi-step reasoning. It supports full, low-effort,...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"ibm-granite","completion_ratio":4.1667,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,ibm-granite","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.25,"openrouter_id":"ibm-granite/granite-4.2-8b","context_length":131072}},{"model_name":"inclusionai/ling-3.0-flash-vl","description":"Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities while adding native visual perception and advanced visual...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.18,"openrouter_id":"inclusionai/ling-3.0-flash-vl","context_length":131072}},{"model_name":"ling-3.0-flash-vl","description":"Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities while adding native visual perception and advanced visual...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.18,"openrouter_id":"inclusionai/ling-3.0-flash-vl","context_length":131072}},{"model_name":"nemotron-3-nano-30b-a3b","description":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"nvidia","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.24,"openrouter_id":"nvidia/nemotron-3-nano-30b-a3b","context_length":262144}},{"model_name":"nova-lite-v1","description":"Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output. Amazon Nova Lite...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"amazon","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.24,"openrouter_id":"amazon/nova-lite-v1","context_length":300000}},{"model_name":"nvidia/nemotron-3-nano-30b-a3b","description":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...","quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"nvidia","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.06,"output_usd_per_m":0.24,"openrouter_id":"nvidia/nemotron-3-nano-30b-a3b","context_length":262144}},{"model_name":"glm-4.7-flash","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,...","quota_type":0,"model_ratio":0.03025,"model_price":0,"owner_by":"glm","completion_ratio":6.6116,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.060500000000000005,"output_usd_per_m":0.39999999999999997,"openrouter_id":"z-ai/glm-4.7-flash","context_length":200000}},{"model_name":"z-ai/glm-4.7-flash","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,...","quota_type":0,"model_ratio":0.03025,"model_price":0,"owner_by":"glm","completion_ratio":6.6116,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.060500000000000005,"output_usd_per_m":0.39999999999999997,"openrouter_id":"z-ai/glm-4.7-flash","context_length":200000}},{"model_name":"qwen/qwen3.5-flash-02-23","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...","quota_type":0,"model_ratio":0.0325,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.065,"output_usd_per_m":0.26,"openrouter_id":"qwen/qwen3.5-flash-02-23","context_length":1000000}},{"model_name":"qwen3.5-flash-02-23","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...","quota_type":0,"model_ratio":0.0325,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.065,"output_usd_per_m":0.26,"openrouter_id":"qwen/qwen3.5-flash-02-23","context_length":1000000}},{"model_name":"~z-ai/glm-latest","description":"This model always redirects to the latest GLM model from Z.ai.","quota_type":0,"model_ratio":0.03495,"model_price":0,"owner_by":"~z-ai","completion_ratio":20.0286,"cache_ratio":0.985694,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~z-ai","grokcode":{"input_usd_per_m":0.0699,"output_usd_per_m":1.4,"openrouter_id":"~z-ai/glm-latest","context_length":1048576}},{"model_name":"glm-5.3","description":"GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...","quota_type":0,"model_ratio":0.03495,"model_price":0,"owner_by":"glm","completion_ratio":20.0286,"cache_ratio":0.985694,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.0699,"output_usd_per_m":1.4,"openrouter_id":"z-ai/glm-5.3","context_length":1048576}},{"model_name":"glm-latest","description":"This model always redirects to the latest GLM model from Z.ai.","quota_type":0,"model_ratio":0.03495,"model_price":0,"owner_by":"~z-ai","completion_ratio":20.0286,"cache_ratio":0.985694,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~z-ai","grokcode":{"input_usd_per_m":0.0699,"output_usd_per_m":1.4,"openrouter_id":"~z-ai/glm-latest","context_length":1048576}},{"model_name":"z-ai/glm-5.3","description":"GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...","quota_type":0,"model_ratio":0.03495,"model_price":0,"owner_by":"glm","completion_ratio":20.0286,"cache_ratio":0.985694,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.0699,"output_usd_per_m":1.4,"openrouter_id":"z-ai/glm-5.3","context_length":1048576}},{"model_name":"microsoft/phi-4","description":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion...","quota_type":0,"model_ratio":0.035,"model_price":0,"owner_by":"microsoft","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,microsoft","grokcode":{"input_usd_per_m":0.07,"output_usd_per_m":0.14,"openrouter_id":"microsoft/phi-4","context_length":16384}},{"model_name":"nemotron-3.5-lightning","description":"NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...","quota_type":0,"model_ratio":0.035,"model_price":0,"owner_by":"nvidia","completion_ratio":2.8571,"cache_ratio":0.571429,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.07,"output_usd_per_m":0.19999999999999998,"openrouter_id":"nvidia/nemotron-3.5-lightning","context_length":262144}},{"model_name":"nvidia/nemotron-3.5-lightning","description":"NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...","quota_type":0,"model_ratio":0.035,"model_price":0,"owner_by":"nvidia","completion_ratio":2.8571,"cache_ratio":0.571429,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.07,"output_usd_per_m":0.19999999999999998,"openrouter_id":"nvidia/nemotron-3.5-lightning","context_length":262144}},{"model_name":"phi-4","description":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion...","quota_type":0,"model_ratio":0.035,"model_price":0,"owner_by":"microsoft","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,microsoft","grokcode":{"input_usd_per_m":0.07,"output_usd_per_m":0.14,"openrouter_id":"microsoft/phi-4","context_length":16384}},{"model_name":"hy-mt2-30b-a3b","description":"Hy-MT2-30B-A3B is Tencent's flagship translation model in the Hy-MT2 family. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and...","quota_type":0,"model_ratio":0.037,"model_price":0,"owner_by":"tencent","completion_ratio":3.9865,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.074,"output_usd_per_m":0.295,"openrouter_id":"tencent/hy-mt2-30b-a3b","context_length":8192}},{"model_name":"hy-mt2-7b","description":"Hy-MT2-7B is a 7B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation.","quota_type":0,"model_ratio":0.037,"model_price":0,"owner_by":"tencent","completion_ratio":3.9865,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.074,"output_usd_per_m":0.295,"openrouter_id":"tencent/hy-mt2-7b","context_length":8192}},{"model_name":"tencent/hy-mt2-30b-a3b","description":"Hy-MT2-30B-A3B is Tencent's flagship translation model in the Hy-MT2 family. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and...","quota_type":0,"model_ratio":0.037,"model_price":0,"owner_by":"tencent","completion_ratio":3.9865,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.074,"output_usd_per_m":0.295,"openrouter_id":"tencent/hy-mt2-30b-a3b","context_length":8192}},{"model_name":"tencent/hy-mt2-7b","description":"Hy-MT2-7B is a 7B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation.","quota_type":0,"model_ratio":0.037,"model_price":0,"owner_by":"tencent","completion_ratio":3.9865,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.074,"output_usd_per_m":0.295,"openrouter_id":"tencent/hy-mt2-7b","context_length":8192}},{"model_name":"bytedance-seed/seed-1.6-flash","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...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.3,"openrouter_id":"bytedance-seed/seed-1.6-flash","context_length":262144}},{"model_name":"GLM-5.3-Flash","description":"智谱高效率原生多模态模型，适合编码、长程 Agent 和成本敏感调用。","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"GLM","completion_ratio":3.3333,"cache_ratio":0.2,"create_cache_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,GLM","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.25,"context_length":null}},{"model_name":"gpt-oss-safeguard-20b","description":"gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.3,"openrouter_id":"openai/gpt-oss-safeguard-20b","context_length":131072}},{"model_name":"inclusionai/ling-3.1-flash","description":"Ling 3.1 Flash is a hybrid reasoning mixture-of-experts model from inclusionAI, with 25B active parameters out of 560B total.","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.22499999999999998,"openrouter_id":"inclusionai/ling-3.1-flash","context_length":1000000}},{"model_name":"ling-3.1-flash","description":"Ling 3.1 Flash is a hybrid reasoning mixture-of-experts model from inclusionAI, with 25B active parameters out of 560B total.","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"inclusionai","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inclusionai","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.22499999999999998,"openrouter_id":"inclusionai/ling-3.1-flash","context_length":1000000}},{"model_name":"minimax-m3.1-flash","description":"MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it delivers significant improvements across multiple benchmarks, particularly in complex agentic tasks, coding, and cybersecurity. It is built for complex,...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"minimax","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.3,"openrouter_id":"minimax/minimax-m3.1-flash","context_length":1048576}},{"model_name":"minimax/minimax-m3.1-flash","description":"MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it delivers significant improvements across multiple benchmarks, particularly in complex agentic tasks, coding, and cybersecurity. It is built for complex,...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"minimax","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.3,"openrouter_id":"minimax/minimax-m3.1-flash","context_length":1048576}},{"model_name":"mistral-small-3.2-24b-instruct","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...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"mistral","completion_ratio":2.6667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.19999999999999998,"openrouter_id":"mistralai/mistral-small-3.2-24b-instruct","context_length":256000}},{"model_name":"mistralai/mistral-small-3.2-24b-instruct","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...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"mistral","completion_ratio":2.6667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.19999999999999998,"openrouter_id":"mistralai/mistral-small-3.2-24b-instruct","context_length":256000}},{"model_name":"nex-agi/nex-n2.5-pro","description":"Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"nex-agi","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nex-agi","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.25,"openrouter_id":"nex-agi/nex-n2.5-pro","context_length":262144}},{"model_name":"nex-n2.5-pro","description":"Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"nex-agi","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nex-agi","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.25,"openrouter_id":"nex-agi/nex-n2.5-pro","context_length":262144}},{"model_name":"openai/gpt-4o-mini:batch","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.3,"openrouter_id":"openai/gpt-4o-mini:batch","context_length":128000}},{"model_name":"openai/gpt-oss-safeguard-20b","description":"gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.3,"openrouter_id":"openai/gpt-oss-safeguard-20b","context_length":131072}},{"model_name":"prism-ml/ternary-bonsai-2-27b","description":"Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window. Ternary compression shrinks...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"prism-ml","completion_ratio":6.6667,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,prism-ml","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.5,"openrouter_id":"prism-ml/ternary-bonsai-2-27b","context_length":262144}},{"model_name":"seed-1.6-flash","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...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.3,"openrouter_id":"bytedance-seed/seed-1.6-flash","context_length":262144}},{"model_name":"ternary-bonsai-2-27b","description":"Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window. Ternary compression shrinks...","quota_type":0,"model_ratio":0.0375,"model_price":0,"owner_by":"prism-ml","completion_ratio":6.6667,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,prism-ml","grokcode":{"input_usd_per_m":0.075,"output_usd_per_m":0.5,"openrouter_id":"prism-ml/ternary-bonsai-2-27b","context_length":262144}},{"model_name":"gemma-4-26b-a4b-it","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...","quota_type":0,"model_ratio":0.03825,"model_price":0,"owner_by":"gemma","completion_ratio":3.3333,"cache_ratio":0.555556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.0765,"output_usd_per_m":0.255,"openrouter_id":"google/gemma-4-26b-a4b-it","context_length":262144}},{"model_name":"google/gemma-4-26b-a4b-it","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...","quota_type":0,"model_ratio":0.03825,"model_price":0,"owner_by":"gemma","completion_ratio":3.3333,"cache_ratio":0.555556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.0765,"output_usd_per_m":0.255,"openrouter_id":"google/gemma-4-26b-a4b-it","context_length":262144}},{"model_name":"gemma-3-27b-it","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,...","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"gemma","completion_ratio":4.375,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.35,"openrouter_id":"google/gemma-3-27b-it","context_length":131072}},{"model_name":"google/gemma-3-27b-it","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,...","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"gemma","completion_ratio":4.375,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.35,"openrouter_id":"google/gemma-3-27b-it","context_length":131072}},{"model_name":"gryphe/mythomax-l2-13b","description":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"gryphe","completion_ratio":1.375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gryphe","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.11,"openrouter_id":"gryphe/mythomax-l2-13b","context_length":8192}},{"model_name":"mythomax-l2-13b","description":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"gryphe","completion_ratio":1.375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gryphe","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.11,"openrouter_id":"gryphe/mythomax-l2-13b","context_length":8192}},{"model_name":"prism-ml/ternary-bonsai-2-9b","description":"Ternary Bonsai 2 9B is PrismML’s compact, open-weight language model for conversational AI, reasoning, and tool use. Built on Qwen3.5, it uses ternary weights packed into 2 bits for efficient...","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"prism-ml","completion_ratio":1.625,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,prism-ml","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.13,"openrouter_id":"prism-ml/ternary-bonsai-2-9b","context_length":262144}},{"model_name":"qwen/qwen3-32b","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...","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"qwen","completion_ratio":3.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.28,"openrouter_id":"qwen/qwen3-32b","context_length":131072}},{"model_name":"qwen3-32b","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...","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"qwen","completion_ratio":3.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.28,"openrouter_id":"qwen/qwen3-32b","context_length":131072}},{"model_name":"ternary-bonsai-2-9b","description":"Ternary Bonsai 2 9B is PrismML’s compact, open-weight language model for conversational AI, reasoning, and tool use. Built on Qwen3.5, it uses ternary weights packed into 2 bits for efficient...","quota_type":0,"model_ratio":0.04,"model_price":0,"owner_by":"prism-ml","completion_ratio":1.625,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,prism-ml","grokcode":{"input_usd_per_m":0.08,"output_usd_per_m":0.13,"openrouter_id":"prism-ml/ternary-bonsai-2-9b","context_length":262144}},{"model_name":"hy3","description":"Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...","quota_type":0,"model_ratio":0.04125,"model_price":0,"owner_by":"tencent","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.0825,"output_usd_per_m":0.33,"openrouter_id":"tencent/hy3","context_length":262144}},{"model_name":"tencent/hy3","description":"Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...","quota_type":0,"model_ratio":0.04125,"model_price":0,"owner_by":"tencent","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.0825,"output_usd_per_m":0.33,"openrouter_id":"tencent/hy3","context_length":262144}},{"model_name":"qwen/qwen3-next-80b-a3b-instruct","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...","quota_type":0,"model_ratio":0.045,"model_price":0,"owner_by":"qwen","completion_ratio":12.2222,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09,"output_usd_per_m":1.1,"openrouter_id":"qwen/qwen3-next-80b-a3b-instruct","context_length":262144}},{"model_name":"qwen3-next-80b-a3b-instruct","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...","quota_type":0,"model_ratio":0.045,"model_price":0,"owner_by":"qwen","completion_ratio":12.2222,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09,"output_usd_per_m":1.1,"openrouter_id":"qwen/qwen3-next-80b-a3b-instruct","context_length":262144}},{"model_name":"deepseek-v4-flash","description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...","quota_type":0,"model_ratio":0.0455,"model_price":0,"owner_by":"deepseek","completion_ratio":2,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.091,"output_usd_per_m":0.182,"openrouter_id":"deepseek/deepseek-v4-flash","context_length":1048576}},{"model_name":"deepseek/deepseek-v4-flash","description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...","quota_type":0,"model_ratio":0.0455,"model_price":0,"owner_by":"deepseek","completion_ratio":2,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.091,"output_usd_per_m":0.182,"openrouter_id":"deepseek/deepseek-v4-flash","context_length":1048576}},{"model_name":"~deepseek/deepseek-pro-latest","description":"This model always redirects to the latest model in the DeepSeek Pro family.","quota_type":0,"model_ratio":0.04635,"model_price":0,"owner_by":"~deepseek","completion_ratio":86.2999,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~deepseek","grokcode":{"input_usd_per_m":0.0927,"output_usd_per_m":8,"openrouter_id":"~deepseek/deepseek-pro-latest","context_length":1048576}},{"model_name":"deepseek-pro-latest","description":"This model always redirects to the latest model in the DeepSeek Pro family.","quota_type":0,"model_ratio":0.04635,"model_price":0,"owner_by":"~deepseek","completion_ratio":86.2999,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~deepseek","grokcode":{"input_usd_per_m":0.0927,"output_usd_per_m":8,"openrouter_id":"~deepseek/deepseek-pro-latest","context_length":1048576}},{"model_name":"~anthropic/claude-haiku-latest","description":"This model always redirects to the latest model in the Claude Haiku family.","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"~anthropic/claude-haiku-latest","context_length":1000000}},{"model_name":"~openai/gpt-luna-latest","description":"This model always redirects to the latest model in the GPT Luna family.","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"~openai/gpt-luna-latest","context_length":1050000}},{"model_name":"anthropic/claude-haiku-5.5","description":"Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"anthropic/claude-haiku-5.5","context_length":1000000}},{"model_name":"bytedance-seed/seed-2.0-mini","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 und","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.39999999999999997,"openrouter_id":"bytedance-seed/seed-2.0-mini","context_length":262144}},{"model_name":"bytedance/ui-tars-1.5-7b","description":"UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games. Built by ByteDance, it builds upon the UI-TARS framework with reinforcement...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"bytedance","completion_ratio":2,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"bytedance/ui-tars-1.5-7b","context_length":128000}},{"model_name":"claude-haiku-5.5","description":"Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"anthropic/claude-haiku-5.5","context_length":1000000}},{"model_name":"claude-haiku-latest","description":"This model always redirects to the latest model in the Claude Haiku family.","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"~anthropic/claude-haiku-latest","context_length":1000000}},{"model_name":"gemini-2.5-flash-lite","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"google","completion_ratio":4,"cache_ratio":0.1,"create_cache_ratio":0.833333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.39999999999999997,"openrouter_id":"google/gemini-2.5-flash-lite","context_length":1048576}},{"model_name":"google/gemini-2.5-flash-lite","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"google","completion_ratio":4,"cache_ratio":0.1,"create_cache_ratio":0.833333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.39999999999999997,"openrouter_id":"google/gemini-2.5-flash-lite","context_length":1048576}},{"model_name":"gpt-4.1-nano","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.39999999999999997,"openrouter_id":"openai/gpt-4.1-nano","context_length":1047576}},{"model_name":"gpt-6-luna","description":"GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"openai/gpt-6-luna","context_length":1050000}},{"model_name":"gpt-6-luna-pro","description":"GPT-6 Luna Pro is the same underlying model as [GPT-6 Luna](https://openrouter.ai/openai/gpt-6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"openai/gpt-6-luna-pro","context_length":1050000}},{"model_name":"gpt-luna-latest","description":"This model always redirects to the latest model in the GPT Luna family.","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"~openai/gpt-luna-latest","context_length":1050000}},{"model_name":"inception/mercury-3","description":"Mercury 3 is a diffusion-based language model from Inception, and the lab's most capable model to date. It is suited for agentic workflows and real-time applications where latency compounds, such...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"inception","completion_ratio":3.5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inception","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.35,"openrouter_id":"inception/mercury-3","context_length":1000000}},{"model_name":"llama-4-scout","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"llama","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"meta-llama/llama-4-scout","context_length":1310720}},{"model_name":"mercury-3","description":"Mercury 3 is a diffusion-based language model from Inception, and the lab's most capable model to date. It is suited for agentic workflows and real-time applications where latency compounds, such...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"inception","completion_ratio":3.5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inception","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.35,"openrouter_id":"inception/mercury-3","context_length":1000000}},{"model_name":"meta-llama/llama-4-scout","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"llama","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"meta-llama/llama-4-scout","context_length":1310720}},{"model_name":"meta/muse-spark-1.2-contributor","description":"Muse Spark 1.2 contributor tier is a reasoning model from Meta designed for developers who want to start building at an even lower cost. It’s meaningfully cheaper than Muse Spark...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"llama","completion_ratio":2,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"meta/muse-spark-1.2-contributor","context_length":1048576}},{"model_name":"meta/muse-spark-1.3-contributor","description":"Muse Spark 1.3 Contributor is the cost-efficient contributor tier of Meta’s multimodal reasoning model for experimentation, learning, and early-stage agentic, multi-agent, and coding workflows. It is designed to track information...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"llama","completion_ratio":2,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"meta/muse-spark-1.3-contributor","context_length":1048576}},{"model_name":"ministral-3b-2512","description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"mistral","completion_ratio":1,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.09999999999999999,"openrouter_id":"mistralai/ministral-3b-2512","context_length":131072}},{"model_name":"mistralai/ministral-3b-2512","description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"mistral","completion_ratio":1,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.09999999999999999,"openrouter_id":"mistralai/ministral-3b-2512","context_length":131072}},{"model_name":"mistralai/voxtral-small-24b-2507","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"mistralai/voxtral-small-24b-2507","context_length":32768}},{"model_name":"muse-spark-1.2-contributor","description":"Muse Spark 1.2 contributor tier is a reasoning model from Meta designed for developers who want to start building at an even lower cost. It’s meaningfully cheaper than Muse Spark...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"llama","completion_ratio":2,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"meta/muse-spark-1.2-contributor","context_length":1048576}},{"model_name":"muse-spark-1.3-contributor","description":"Muse Spark 1.3 Contributor is the cost-efficient contributor tier of Meta’s multimodal reasoning model for experimentation, learning, and early-stage agentic, multi-agent, and coding workflows. It is designed to track information...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"llama","completion_ratio":2,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"meta/muse-spark-1.3-contributor","context_length":1048576}},{"model_name":"openai/gpt-4.1-nano","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.39999999999999997,"openrouter_id":"openai/gpt-4.1-nano","context_length":1047576}},{"model_name":"openai/gpt-5.4-nano:batch","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":6.25,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.625,"openrouter_id":"openai/gpt-5.4-nano:batch","context_length":400000}},{"model_name":"openai/gpt-5.6-luna-pro:batch","description":"GPT-5.6 Luna Pro is the same underlying model as [GPT-5.6 Luna](https://openrouter.ai/openai/gpt-5.6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.6,"openrouter_id":"openai/gpt-5.6-luna-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-5.6-luna:batch","description":"GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.6,"openrouter_id":"openai/gpt-5.6-luna:batch","context_length":1050000}},{"model_name":"openai/gpt-6-luna","description":"GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"openai/gpt-6-luna","context_length":1050000}},{"model_name":"openai/gpt-6-luna-pro","description":"GPT-6 Luna Pro is the same underlying model as [GPT-6 Luna](https://openrouter.ai/openai/gpt-6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.5,"openrouter_id":"openai/gpt-6-luna-pro","context_length":1050000}},{"model_name":"qwen-2.5-7b-instruct","description":"Qwen2.5 7B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"qwen","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"qwen/qwen-2.5-7b-instruct","context_length":32768}},{"model_name":"qwen/qwen-2.5-7b-instruct","description":"Qwen2.5 7B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"qwen","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"qwen/qwen-2.5-7b-instruct","context_length":32768}},{"model_name":"qwen/qwen3-30b-a3b-instruct-2507","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"qwen/qwen3-30b-a3b-instruct-2507","context_length":262144}},{"model_name":"qwen/qwen3.6-35b-a3b","description":"Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"qwen","completion_ratio":9,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.8999999999999999,"openrouter_id":"qwen/qwen3.6-35b-a3b","context_length":262144}},{"model_name":"qwen3-30b-a3b-instruct-2507","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"qwen/qwen3-30b-a3b-instruct-2507","context_length":262144}},{"model_name":"qwen3.6-35b-a3b","description":"Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"qwen","completion_ratio":9,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.8999999999999999,"openrouter_id":"qwen/qwen3.6-35b-a3b","context_length":262144}},{"model_name":"reka-edge","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,...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"rekaai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,rekaai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.09999999999999999,"openrouter_id":"rekaai/reka-edge","context_length":16384}},{"model_name":"reka-flash-3","description":"Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka. It excels at general chat, coding tasks, instruction-following, and function calling. Featuring a...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"rekaai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,rekaai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"rekaai/reka-flash-3","context_length":65536}},{"model_name":"rekaai/reka-edge","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,...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"rekaai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,rekaai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.09999999999999999,"openrouter_id":"rekaai/reka-edge","context_length":16384}},{"model_name":"rekaai/reka-flash-3","description":"Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka. It excels at general chat, coding tasks, instruction-following, and function calling. Featuring a...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"rekaai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,rekaai","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"rekaai/reka-flash-3","context_length":65536}},{"model_name":"seed-2.0-mini","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 und","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.39999999999999997,"openrouter_id":"bytedance-seed/seed-2.0-mini","context_length":262144}},{"model_name":"step-3.5-flash","description":"Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"stepfun","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stepfun","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"stepfun/step-3.5-flash","context_length":262144}},{"model_name":"stepfun/step-3.5-flash","description":"Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"stepfun","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stepfun","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"stepfun/step-3.5-flash","context_length":262144}},{"model_name":"ui-tars-1.5-7b","description":"UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games. Built by ByteDance, it builds upon the UI-TARS framework with reinforcement...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"bytedance","completion_ratio":2,"cache_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.19999999999999998,"openrouter_id":"bytedance/ui-tars-1.5-7b","context_length":128000}},{"model_name":"voxtral-small-24b-2507","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...","quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.09999999999999999,"output_usd_per_m":0.3,"openrouter_id":"mistralai/voxtral-small-24b-2507","context_length":32768}},{"model_name":"deepseek-v4-pro-0813","description":"DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.","quota_type":0,"model_ratio":0.0515,"model_price":0,"owner_by":"deepseek","completion_ratio":39.9029,"cache_ratio":0.951456,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.103,"output_usd_per_m":4.109999999999999,"openrouter_id":"deepseek/deepseek-v4-pro-0813","context_length":1048576}},{"model_name":"deepseek/deepseek-v4-pro-0813","description":"DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.","quota_type":0,"model_ratio":0.0515,"model_price":0,"owner_by":"deepseek","completion_ratio":39.9029,"cache_ratio":0.951456,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.103,"output_usd_per_m":4.109999999999999,"openrouter_id":"deepseek/deepseek-v4-pro-0813","context_length":1048576}},{"model_name":"deepseek/deepseek-v4.1-flash:batch","description":"DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...","quota_type":0,"model_ratio":0.056,"model_price":0,"owner_by":"deepseek","completion_ratio":3,"cache_ratio":0.03,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.112,"output_usd_per_m":0.33599999999999997,"openrouter_id":"deepseek/deepseek-v4.1-flash:batch","context_length":1048576}},{"model_name":"qwen/qwen3-14b","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...","quota_type":0,"model_ratio":0.06,"model_price":0,"owner_by":"qwen","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.12,"output_usd_per_m":0.24,"openrouter_id":"qwen/qwen3-14b","context_length":40960}},{"model_name":"qwen/qwen3-30b-a3b","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...","quota_type":0,"model_ratio":0.06,"model_price":0,"owner_by":"qwen","completion_ratio":4.1667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.12,"output_usd_per_m":0.5,"openrouter_id":"qwen/qwen3-30b-a3b","context_length":40960}},{"model_name":"qwen/qwen3-coder-next","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...","quota_type":0,"model_ratio":0.06,"model_price":0,"owner_by":"qwen","completion_ratio":6.6667,"cache_ratio":0.583333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.12,"output_usd_per_m":0.7999999999999999,"openrouter_id":"qwen/qwen3-coder-next","context_length":262144}},{"model_name":"qwen3-14b","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...","quota_type":0,"model_ratio":0.06,"model_price":0,"owner_by":"qwen","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.12,"output_usd_per_m":0.24,"openrouter_id":"qwen/qwen3-14b","context_length":40960}},{"model_name":"qwen3-30b-a3b","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...","quota_type":0,"model_ratio":0.06,"model_price":0,"owner_by":"qwen","completion_ratio":4.1667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.12,"output_usd_per_m":0.5,"openrouter_id":"qwen/qwen3-30b-a3b","context_length":40960}},{"model_name":"qwen3-coder-next","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...","quota_type":0,"model_ratio":0.06,"model_price":0,"owner_by":"qwen","completion_ratio":6.6667,"cache_ratio":0.583333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.12,"output_usd_per_m":0.7999999999999999,"openrouter_id":"qwen/qwen3-coder-next","context_length":262144}},{"model_name":"google/gemini-3.1-flash-lite:batch","description":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads. It supports text, image, video, audio, and PDF inputs, and is designed for lightweight agentic...","quota_type":0,"model_ratio":0.0625,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.125,"output_usd_per_m":0.75,"openrouter_id":"google/gemini-3.1-flash-lite:batch","context_length":1048576}},{"model_name":"openai/gpt-5-mini:batch","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....","quota_type":0,"model_ratio":0.0625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.125,"output_usd_per_m":1,"openrouter_id":"openai/gpt-5-mini:batch","context_length":400000}},{"model_name":"glm-4.5-air","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...","quota_type":0,"model_ratio":0.065,"model_price":0,"owner_by":"glm","completion_ratio":6.5385,"cache_ratio":0.192308,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.13,"output_usd_per_m":0.85,"openrouter_id":"z-ai/glm-4.5-air","context_length":131072}},{"model_name":"z-ai/glm-4.5-air","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...","quota_type":0,"model_ratio":0.065,"model_price":0,"owner_by":"glm","completion_ratio":6.5385,"cache_ratio":0.192308,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.13,"output_usd_per_m":0.85,"openrouter_id":"z-ai/glm-4.5-air","context_length":131072}},{"model_name":"hunyuan-a13b-instruct","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...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"tencent","completion_ratio":4.0714,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":0.5700000000000001,"openrouter_id":"tencent/hunyuan-a13b-instruct","context_length":131072}},{"model_name":"MiMo V2.5","description":"Xiaomi MiMo V2.5 系列均衡基座模型，官方强调全模态感知、Agent 能力和 1M 上下文，适合成本与能力平衡的中文模型调用。","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"MiMo","completion_ratio":2,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,MiMo","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":0.28,"context_length":null}},{"model_name":"mimo-v2.6-flash","description":"MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"xiaomi","completion_ratio":2,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":0.28,"openrouter_id":"xiaomi/mimo-v2.6-flash","context_length":1050000}},{"model_name":"qwen/qwen3-235b-a22b-2507","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,...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"qwen","completion_ratio":5.7143,"cache_ratio":0.357143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":0.7999999999999999,"openrouter_id":"qwen/qwen3-235b-a22b-2507","context_length":262144}},{"model_name":"qwen/qwen3.5-35b-a3b","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...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"qwen","completion_ratio":7.1429,"cache_ratio":0.357143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":1,"openrouter_id":"qwen/qwen3.5-35b-a3b","context_length":262144}},{"model_name":"qwen3-235b-a22b-2507","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,...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"qwen","completion_ratio":5.7143,"cache_ratio":0.357143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":0.7999999999999999,"openrouter_id":"qwen/qwen3-235b-a22b-2507","context_length":262144}},{"model_name":"qwen3.5-35b-a3b","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...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"qwen","completion_ratio":7.1429,"cache_ratio":0.357143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":1,"openrouter_id":"qwen/qwen3.5-35b-a3b","context_length":262144}},{"model_name":"tencent/hunyuan-a13b-instruct","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...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"tencent","completion_ratio":4.0714,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":0.5700000000000001,"openrouter_id":"tencent/hunyuan-a13b-instruct","context_length":131072}},{"model_name":"xiaomi/mimo-v2.6-flash","description":"MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for...","quota_type":0,"model_ratio":0.07,"model_price":0,"owner_by":"xiaomi","completion_ratio":2,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.14,"output_usd_per_m":0.28,"openrouter_id":"xiaomi/mimo-v2.6-flash","context_length":1050000}},{"model_name":"cohere/command-r-08-2024","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"cohere/command-r-08-2024","context_length":128000}},{"model_name":"command-r-08-2024","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"cohere/command-r-08-2024","context_length":128000}},{"model_name":"deepseek-v4.1-flash-1017","description":"DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name. V4.1 Flash is a sparse mixture-of-experts model...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"deepseek","completion_ratio":4,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"deepseek/deepseek-v4.1-flash-1017","context_length":1048576}},{"model_name":"deepseek/deepseek-v4.1-flash-1017","description":"DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name. V4.1 Flash is a sparse mixture-of-experts model...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"deepseek","completion_ratio":4,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"deepseek/deepseek-v4.1-flash-1017","context_length":1048576}},{"model_name":"flash","description":"低峰 · 未命中缓存。模型名 deepseek-flash，版本 DeepSeek-V4.1-Flash。输入 0.15 美元/M，输出 0.6 美元/M。","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"deepseek","completion_ratio":4,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,deepseek","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"context_length":null}},{"model_name":"glm-5.3-flash","description":"GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"glm","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.5,"openrouter_id":"z-ai/glm-5.3-flash","context_length":1048576}},{"model_name":"google/gemini-3.5-flash-lite:batch","description":"Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"google","completion_ratio":8.3333,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":1.25,"openrouter_id":"google/gemini-3.5-flash-lite:batch","context_length":1048576}},{"model_name":"gpt-4o-mini","description":"gpt-4o-mini","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"context_length":null}},{"model_name":"gpt-4o-mini-2024-07-18","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"openai/gpt-4o-mini-2024-07-18","context_length":128000}},{"model_name":"gpt-oss-120b","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"openai/gpt-oss-120b","context_length":131072}},{"model_name":"ministral-8b-2512","description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"mistral","completion_ratio":1,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.15,"openrouter_id":"mistralai/ministral-8b-2512","context_length":262144}},{"model_name":"mistral-small-2603","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"mistral","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"mistralai/mistral-small-2603","context_length":262144}},{"model_name":"mistralai/ministral-8b-2512","description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"mistral","completion_ratio":1,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.15,"openrouter_id":"mistralai/ministral-8b-2512","context_length":262144}},{"model_name":"mistralai/mistral-small-2603","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"mistral","completion_ratio":4,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"mistralai/mistral-small-2603","context_length":262144}},{"model_name":"nemotron-3.5-super-vl","description":"NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"nvidia","completion_ratio":3.3333,"cache_ratio":0.133333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.5,"openrouter_id":"nvidia/nemotron-3.5-super-vl","context_length":1048576}},{"model_name":"nvidia/nemotron-3.5-super-vl","description":"NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"nvidia","completion_ratio":3.3333,"cache_ratio":0.133333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.5,"openrouter_id":"nvidia/nemotron-3.5-super-vl","context_length":1048576}},{"model_name":"openai/gpt-4o-mini","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"openai/gpt-4o-mini","context_length":128000}},{"model_name":"openai/gpt-4o-mini-2024-07-18","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"openai/gpt-4o-mini-2024-07-18","context_length":128000}},{"model_name":"openai/gpt-oss-120b","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"openai/gpt-oss-120b","context_length":131072}},{"model_name":"perceptron-mk1","description":"Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces detailed visual understanding...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"perceptron","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perceptron","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":1.5,"openrouter_id":"perceptron/perceptron-mk1","context_length":32768}},{"model_name":"perceptron-mk1.5","description":"Perceptron Mk1.5 is Perceptron's embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers with text plus optional structured annotations: points, boxes, polygons, tracks,...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"perceptron","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perceptron","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":1.5,"openrouter_id":"perceptron/perceptron-mk1.5","context_length":36864}},{"model_name":"perceptron/perceptron-mk1","description":"Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces detailed visual understanding...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"perceptron","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perceptron","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":1.5,"openrouter_id":"perceptron/perceptron-mk1","context_length":32768}},{"model_name":"perceptron/perceptron-mk1.5","description":"Perceptron Mk1.5 is Perceptron's embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers with text plus optional structured annotations: points, boxes, polygons, tracks,...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"perceptron","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perceptron","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":1.5,"openrouter_id":"perceptron/perceptron-mk1.5","context_length":36864}},{"model_name":"qwen/qwen3-coder-30b-a3b-instruct","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"qwen/qwen3-coder-30b-a3b-instruct","context_length":262144}},{"model_name":"qwen/qwen3-vl-30b-a3b-instruct","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"qwen/qwen3-vl-30b-a3b-instruct","context_length":262144}},{"model_name":"qwen/qwen3.8-27b","description":"Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":12.5,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":1.875,"openrouter_id":"qwen/qwen3.8-27b","context_length":1024000}},{"model_name":"qwen/qwen3.8-flash","description":"Qwen3.8 Flash is a multimodal reasoning model from Alibaba. It is suited for coding assistance, agentic workflows, visual understanding, document and codebase analysis, desktop interaction, chart analysis, and long-video analysis.","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":3.1333,"cache_ratio":0.106667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.47,"openrouter_id":"qwen/qwen3.8-flash","context_length":1000000}},{"model_name":"qwen/qwen3.8-omni-flash","description":"Qwen3.8 Omni Flash is an omni-modal reasoning model from Alibaba, the first Qwen model built around agentic capabilities with native audio-video understanding. It is suited for audio-video analysis and summarization,...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":3.1333,"cache_ratio":0.106667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.47,"openrouter_id":"qwen/qwen3.8-omni-flash","context_length":1000000}},{"model_name":"qwen3-coder-30b-a3b-instruct","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"qwen/qwen3-coder-30b-a3b-instruct","context_length":262144}},{"model_name":"qwen3-vl-30b-a3b-instruct","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...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"qwen/qwen3-vl-30b-a3b-instruct","context_length":262144}},{"model_name":"Qwen3.8 Flash","description":"Qwen3.8 高效率多模态模型，适合视觉推理、文档处理和成本敏感的 Agent 任务。","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"Qwen","completion_ratio":3.1333,"cache_ratio":0.106667,"create_cache_ratio":1.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.47,"context_length":null}},{"model_name":"qwen3.8-27b","description":"Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":12.5,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":1.875,"openrouter_id":"qwen/qwen3.8-27b","context_length":1024000}},{"model_name":"qwen3.8-flash","description":"Qwen3.8 Flash is a multimodal reasoning model from Alibaba. It is suited for coding assistance, agentic workflows, visual understanding, document and codebase analysis, desktop interaction, chart analysis, and long-video analysis.","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":3.1333,"cache_ratio":0.106667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.47,"openrouter_id":"qwen/qwen3.8-flash","context_length":1000000}},{"model_name":"qwen3.8-omni-flash","description":"Qwen3.8 Omni Flash is an omni-modal reasoning model from Alibaba, the first Qwen model built around agentic capabilities with native audio-video understanding. It is suited for audio-video analysis and summarization,...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"qwen","completion_ratio":3.1333,"cache_ratio":0.106667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.47,"openrouter_id":"qwen/qwen3.8-omni-flash","context_length":1000000}},{"model_name":"solar-pro4","description":"Solar Pro 4 is Upstage's cost-efficient large language model, featuring a 524K context window. It is built for long-horizon tasks and agentic workflows, with strong capabilities in office productivity, document-intensive...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"upstage","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,upstage","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"upstage/solar-pro4","context_length":524288}},{"model_name":"upstage/solar-pro4","description":"Solar Pro 4 is Upstage's cost-efficient large language model, featuring a 524K context window. It is built for long-horizon tasks and agentic workflows, with strong capabilities in office productivity, document-intensive...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"upstage","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,upstage","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.6,"openrouter_id":"upstage/solar-pro4","context_length":524288}},{"model_name":"z-ai/glm-5.3-flash","description":"GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...","quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"glm","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.15,"output_usd_per_m":0.5,"openrouter_id":"z-ai/glm-5.3-flash","context_length":1048576}},{"model_name":"~moonshotai/kimi-latest","description":"This model always redirects to the latest model in the Kimi family.","quota_type":0,"model_ratio":0.07695,"model_price":0,"owner_by":"~moonshotai","completion_ratio":65.9448,"cache_ratio":0.974659,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~moonshotai","grokcode":{"input_usd_per_m":0.15389999999999998,"output_usd_per_m":10.148900000000001,"openrouter_id":"~moonshotai/kimi-latest","context_length":1048576}},{"model_name":"kimi-latest","description":"This model always redirects to the latest model in the Kimi family.","quota_type":0,"model_ratio":0.07695,"model_price":0,"owner_by":"~moonshotai","completion_ratio":65.9448,"cache_ratio":0.974659,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~moonshotai","grokcode":{"input_usd_per_m":0.15389999999999998,"output_usd_per_m":10.148900000000001,"openrouter_id":"~moonshotai/kimi-latest","context_length":1048576}},{"model_name":"qwen/qwen3.5-9b","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...","quota_type":0,"model_ratio":0.085,"model_price":0,"owner_by":"qwen","completion_ratio":1.4706,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.16999999999999998,"output_usd_per_m":0.25,"openrouter_id":"qwen/qwen3.5-9b","context_length":262144}},{"model_name":"qwen3.5-9b","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...","quota_type":0,"model_ratio":0.085,"model_price":0,"owner_by":"qwen","completion_ratio":1.4706,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.16999999999999998,"output_usd_per_m":0.25,"openrouter_id":"qwen/qwen3.5-9b","context_length":262144}},{"model_name":"hy3-preview","description":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to...","quota_type":0,"model_ratio":0.09,"model_price":0,"owner_by":"tencent","completion_ratio":3.3333,"cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.18,"output_usd_per_m":0.6,"openrouter_id":"tencent/hy3-preview","context_length":262144}},{"model_name":"llama-guard-4-12b","description":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM...","quota_type":0,"model_ratio":0.09,"model_price":0,"owner_by":"llama","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.18,"output_usd_per_m":0.18,"openrouter_id":"meta-llama/llama-guard-4-12b","context_length":163840}},{"model_name":"meta-llama/llama-guard-4-12b","description":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM...","quota_type":0,"model_ratio":0.09,"model_price":0,"owner_by":"llama","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.18,"output_usd_per_m":0.18,"openrouter_id":"meta-llama/llama-guard-4-12b","context_length":163840}},{"model_name":"tencent/hy3-preview","description":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to...","quota_type":0,"model_ratio":0.09,"model_price":0,"owner_by":"tencent","completion_ratio":3.3333,"cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.18,"output_usd_per_m":0.6,"openrouter_id":"tencent/hy3-preview","context_length":262144}},{"model_name":"qwen/qwen3.6-flash","description":"Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series. It supports text, image, and video input with a 1M token context window. Tiered pricing kicks in...","quota_type":0,"model_ratio":0.09375,"model_price":0,"owner_by":"qwen","completion_ratio":6,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.1875,"output_usd_per_m":1.125,"openrouter_id":"qwen/qwen3.6-flash","context_length":1000000}},{"model_name":"qwen3.6-flash","description":"Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series. It supports text, image, and video input with a 1M token context window. Tiered pricing kicks in...","quota_type":0,"model_ratio":0.09375,"model_price":0,"owner_by":"qwen","completion_ratio":6,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.1875,"output_usd_per_m":1.125,"openrouter_id":"qwen/qwen3.6-flash","context_length":1000000}},{"model_name":"kimi-k3","description":"Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...","quota_type":0,"model_ratio":0.096,"model_price":0,"owner_by":"kimi","completion_ratio":83.3333,"cache_ratio":0.947917,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.192,"output_usd_per_m":16,"openrouter_id":"moonshotai/kimi-k3","context_length":1048576}},{"model_name":"moonshotai/kimi-k3","description":"Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...","quota_type":0,"model_ratio":0.096,"model_price":0,"owner_by":"kimi","completion_ratio":83.3333,"cache_ratio":0.947917,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.192,"output_usd_per_m":16,"openrouter_id":"moonshotai/kimi-k3","context_length":1048576}},{"model_name":"qwen/qwen3-coder-flash","description":"Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus. It is a powerful coding agent model specializing in autonomous programming via tool calling...","quota_type":0,"model_ratio":0.0975,"model_price":0,"owner_by":"qwen","completion_ratio":5,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.195,"output_usd_per_m":0.975,"openrouter_id":"qwen/qwen3-coder-flash","context_length":1000000}},{"model_name":"qwen3-coder-flash","description":"Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus. It is a powerful coding agent model specializing in autonomous programming via tool calling...","quota_type":0,"model_ratio":0.0975,"model_price":0,"owner_by":"qwen","completion_ratio":5,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.195,"output_usd_per_m":0.975,"openrouter_id":"qwen/qwen3-coder-flash","context_length":1000000}},{"model_name":"cognitivecomputations/dolphin-mistral-24b-venice-edition","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"cognitivecomputations","completion_ratio":4.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cognitivecomputations","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.8999999999999999,"openrouter_id":"cognitivecomputations/dolphin-mistral-24b-venice-edition","context_length":128000}},{"model_name":"dolphin-mistral-24b-venice-edition","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"cognitivecomputations","completion_ratio":4.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cognitivecomputations","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.8999999999999999,"openrouter_id":"cognitivecomputations/dolphin-mistral-24b-venice-edition","context_length":128000}},{"model_name":"gemma-4-31b-it","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"gemma","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.39999999999999997,"openrouter_id":"google/gemma-4-31b-it","context_length":262144}},{"model_name":"google/gemma-4-31b-it","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"gemma","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.39999999999999997,"openrouter_id":"google/gemma-4-31b-it","context_length":262144}},{"model_name":"gpt-5.6-luna","description":"GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":1.2,"openrouter_id":"openai/gpt-5.6-luna","context_length":1050000}},{"model_name":"gpt-5.6-luna-pro","description":"GPT-5.6 Luna Pro is the same underlying model as [GPT-5.6 Luna](https://openrouter.ai/openai/gpt-5.6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":1.2,"openrouter_id":"openai/gpt-5.6-luna-pro","context_length":1050000}},{"model_name":"ministral-14b-2512","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"mistral","completion_ratio":1,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.19999999999999998,"openrouter_id":"mistralai/ministral-14b-2512","context_length":262144}},{"model_name":"mistral-saba","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.6,"openrouter_id":"mistralai/mistral-saba","context_length":32768}},{"model_name":"mistralai/ministral-14b-2512","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"mistral","completion_ratio":1,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.19999999999999998,"openrouter_id":"mistralai/ministral-14b-2512","context_length":262144}},{"model_name":"mistralai/mistral-saba","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.6,"openrouter_id":"mistralai/mistral-saba","context_length":32768}},{"model_name":"nemotron-3.5-content-safety","description":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It moderates both inputs to and responses from LLMs and VLMs, accepting...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.19999999999999998,"openrouter_id":"nvidia/nemotron-3.5-content-safety","context_length":131072}},{"model_name":"nvidia/nemotron-3.5-content-safety","description":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It moderates both inputs to and responses from LLMs and VLMs, accepting...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.19999999999999998,"openrouter_id":"nvidia/nemotron-3.5-content-safety","context_length":131072}},{"model_name":"openai/gpt-4.1-mini:batch","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.7999999999999999,"openrouter_id":"openai/gpt-4.1-mini:batch","context_length":1047576}},{"model_name":"openai/gpt-5.4-nano","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...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"openai","completion_ratio":6.25,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":1.25,"openrouter_id":"openai/gpt-5.4-nano","context_length":400000}},{"model_name":"openai/gpt-5.6-luna","description":"GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":1.2,"openrouter_id":"openai/gpt-5.6-luna","context_length":1050000}},{"model_name":"openai/gpt-5.6-luna-pro","description":"GPT-5.6 Luna Pro is the same underlying model as [GPT-5.6 Luna](https://openrouter.ai/openai/gpt-5.6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":1.2,"openrouter_id":"openai/gpt-5.6-luna-pro","context_length":1050000}},{"model_name":"ricursiveai/ripple-65b-preview","description":"Ripple-65B-Preview by ricursiveAI, a 65B model with native tool calling, served in BF16.","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"ricursiveai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,ricursiveai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.6,"openrouter_id":"ricursiveai/ripple-65b-preview","context_length":1048576}},{"model_name":"ripple-65b-preview","description":"Ripple-65B-Preview by ricursiveAI, a 65B model with native tool calling, served in BF16.","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"ricursiveai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,ricursiveai","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":0.6,"openrouter_id":"ricursiveai/ripple-65b-preview","context_length":1048576}},{"model_name":"step-3.7-flash","description":"Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"stepfun","completion_ratio":5.75,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stepfun","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":1.15,"openrouter_id":"stepfun/step-3.7-flash","context_length":262144}},{"model_name":"stepfun/step-3.7-flash","description":"Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters...","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"stepfun","completion_ratio":5.75,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stepfun","grokcode":{"input_usd_per_m":0.19999999999999998,"output_usd_per_m":1.15,"openrouter_id":"stepfun/step-3.7-flash","context_length":262144}},{"model_name":"GPT-5.4 Nano","description":"OpenAI 极低成本模型，适合高频分类、短文本处理、批量标签和非关键路径自动化。","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"OpenAI","completion_ratio":6.25,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,OpenAI","grokcode":{"input_usd_per_m":0.2,"output_usd_per_m":1.25,"context_length":null}},{"model_name":"gpt-5.4-nano","description":"gpt-5.4-nano","quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"openai","completion_ratio":6.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":0.2,"output_usd_per_m":1.25,"context_length":null}},{"model_name":"minimax-m2.7","description":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement. Built to actively participate in its own evolution, M2.7 integrates advanced agentic capabilities through multi-agent...","quota_type":0,"model_ratio":0.105,"model_price":0,"owner_by":"minimax","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.21,"output_usd_per_m":0.84,"openrouter_id":"minimax/minimax-m2.7","context_length":204800}},{"model_name":"minimax/minimax-m2.7","description":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement. Built to actively participate in its own evolution, M2.7 integrates advanced agentic capabilities through multi-agent...","quota_type":0,"model_ratio":0.105,"model_price":0,"owner_by":"minimax","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.21,"output_usd_per_m":0.84,"openrouter_id":"minimax/minimax-m2.7","context_length":204800}},{"model_name":"qwen/qwen3-vl-235b-a22b-instruct","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...","quota_type":0,"model_ratio":0.105,"model_price":0,"owner_by":"qwen","completion_ratio":9.0476,"cache_ratio":0.47619,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.21,"output_usd_per_m":1.9,"openrouter_id":"qwen/qwen3-vl-235b-a22b-instruct","context_length":262144}},{"model_name":"qwen3-vl-235b-a22b-instruct","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...","quota_type":0,"model_ratio":0.105,"model_price":0,"owner_by":"qwen","completion_ratio":9.0476,"cache_ratio":0.47619,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.21,"output_usd_per_m":1.9,"openrouter_id":"qwen/qwen3-vl-235b-a22b-instruct","context_length":262144}},{"model_name":"deepseek-v4-flash-vision-exp","description":"DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of [DeepSeek V4 Flash 0731](https://openrouter.ai/deepseek/deepseek-v4-flash-0731) from DeepSeek, adding image understanding while matching the base model on text capabilities including agents,...","quota_type":0,"model_ratio":0.1078,"model_price":0,"owner_by":"deepseek","completion_ratio":3,"cache_ratio":0.031818,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.21559999999999999,"output_usd_per_m":0.6468,"openrouter_id":"deepseek/deepseek-v4-flash-vision-exp","context_length":1048576}},{"model_name":"deepseek/deepseek-v4-flash-vision-exp","description":"DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of [DeepSeek V4 Flash 0731](https://openrouter.ai/deepseek/deepseek-v4-flash-0731) from DeepSeek, adding image understanding while matching the base model on text capabilities including agents,...","quota_type":0,"model_ratio":0.1078,"model_price":0,"owner_by":"deepseek","completion_ratio":3,"cache_ratio":0.031818,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.21559999999999999,"output_usd_per_m":0.6468,"openrouter_id":"deepseek/deepseek-v4-flash-vision-exp","context_length":1048576}},{"model_name":"qwen/qwen3.7-plus:batch","description":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its...","quota_type":0,"model_ratio":0.115,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.22999999999999998,"output_usd_per_m":0.9199999999999999,"openrouter_id":"qwen/qwen3.7-plus:batch","context_length":262144}},{"model_name":"bytedance-seed/seed-1.6","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.","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"bytedance-seed/seed-1.6","context_length":262144}},{"model_name":"bytedance-seed/seed-2.0-lite","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"bytedance-seed/seed-2.0-lite","context_length":262144}},{"model_name":"deepseek-chat-v3.1","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"deepseek","completion_ratio":3.8,"cache_ratio":0.52,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":0.95,"openrouter_id":"deepseek/deepseek-chat-v3.1","context_length":163840}},{"model_name":"deepseek/deepseek-chat-v3.1","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"deepseek","completion_ratio":3.8,"cache_ratio":0.52,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":0.95,"openrouter_id":"deepseek/deepseek-chat-v3.1","context_length":163840}},{"model_name":"Gemini 3.1 Flash-Lite","description":"Google Gemini 低成本高速模型，适合批量文本、多模态轻任务和成本敏感调用。","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"Gemini","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Gemini","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"context_length":null}},{"model_name":"gemini-3.1-flash-lite","description":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads. It supports text, image, video, audio, and PDF inputs, and is designed for lightweight agentic...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"openrouter_id":"google/gemini-3.1-flash-lite","context_length":1048576}},{"model_name":"gemini-3.1-flash-lite-image","description":"Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) is Google's fastest, most cost-efficient Gemini image model, built for high-velocity developer pipelines and rapid-fire visual exploration. It delivers text-to-image generation...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"openrouter_id":"google/gemini-3.1-flash-lite-image","context_length":65536}},{"model_name":"gemini-3.1-flash-lite-preview","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"openrouter_id":"google/gemini-3.1-flash-lite-preview","context_length":1048576}},{"model_name":"google/gemini-3-flash-preview:batch","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"openrouter_id":"google/gemini-3-flash-preview:batch","context_length":1048576}},{"model_name":"google/gemini-3.1-flash-lite","description":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads. It supports text, image, video, audio, and PDF inputs, and is designed for lightweight agentic...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"openrouter_id":"google/gemini-3.1-flash-lite","context_length":1048576}},{"model_name":"google/gemini-3.1-flash-lite-image","description":"Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) is Google's fastest, most cost-efficient Gemini image model, built for high-velocity developer pipelines and rapid-fire visual exploration. It delivers text-to-image generation...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"openrouter_id":"google/gemini-3.1-flash-lite-image","context_length":65536}},{"model_name":"google/gemini-3.1-flash-lite-preview","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":1.5,"openrouter_id":"google/gemini-3.1-flash-lite-preview","context_length":1048576}},{"model_name":"gpt-5-mini","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....","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"openai/gpt-5-mini","context_length":400000}},{"model_name":"gpt-5.1-codex-mini","description":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"openai/gpt-5.1-codex-mini","context_length":400000}},{"model_name":"inception/mercury-2","description":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM). Instead of generating tokens sequentially, Mercury 2 produces and refines multiple tokens in parallel, achieving...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"inception","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inception","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":0.75,"openrouter_id":"inception/mercury-2","context_length":128000}},{"model_name":"mercury-2","description":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM). Instead of generating tokens sequentially, Mercury 2 produces and refines multiple tokens in parallel, achieving...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"inception","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,inception","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":0.75,"openrouter_id":"inception/mercury-2","context_length":128000}},{"model_name":"Nano Banana Lite","description":"Google Gemini 3.1 Flash Lite Image 轻量图片生成模型，适合预算敏感的图片生成任务。","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"图片生成","completion_ratio":0.1344,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,图片生成","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":0.0336,"context_length":null}},{"model_name":"openai/gpt-5-mini","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....","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"openai/gpt-5-mini","context_length":400000}},{"model_name":"openai/gpt-5.1-codex-mini","description":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"openai/gpt-5.1-codex-mini","context_length":400000}},{"model_name":"qwen/qwen3-vl-8b-instruct","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.48,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":0.75,"openrouter_id":"qwen/qwen3-vl-8b-instruct","context_length":262144}},{"model_name":"qwen3-vl-8b-instruct","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.48,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":0.75,"openrouter_id":"qwen/qwen3-vl-8b-instruct","context_length":262144}},{"model_name":"seed-1.6","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.","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"bytedance-seed/seed-1.6","context_length":262144}},{"model_name":"seed-2.0-lite","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...","quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.25,"output_usd_per_m":2,"openrouter_id":"bytedance-seed/seed-2.0-lite","context_length":262144}},{"model_name":"deepseek/deepseek-chat","description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...","quota_type":0,"model_ratio":0.1287,"model_price":0,"owner_by":"deepseek","completion_ratio":3.9965,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.2574,"output_usd_per_m":1.0287,"openrouter_id":"deepseek/deepseek-chat","context_length":163840}},{"model_name":"deepseek-v3.2","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"deepseek","completion_ratio":1.4615,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":0.38,"openrouter_id":"deepseek/deepseek-v3.2","context_length":163840}},{"model_name":"deepseek/deepseek-v3.2","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"deepseek","completion_ratio":1.4615,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":0.38,"openrouter_id":"deepseek/deepseek-v3.2","context_length":163840}},{"model_name":"qwen-plus","description":"Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination.","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":0.78,"openrouter_id":"qwen/qwen-plus","context_length":1000000}},{"model_name":"qwen/qwen-plus","description":"Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination.","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":0.78,"openrouter_id":"qwen/qwen-plus","context_length":1000000}},{"model_name":"qwen/qwen3.5-122b-a10b","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":2.08,"openrouter_id":"qwen/qwen3.5-122b-a10b","context_length":262144}},{"model_name":"qwen/qwen3.5-27b","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":2.6,"openrouter_id":"qwen/qwen3.5-27b","context_length":262144}},{"model_name":"qwen/qwen3.5-plus-02-15","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":1.56,"openrouter_id":"qwen/qwen3.5-plus-02-15","context_length":1000000}},{"model_name":"qwen3.5-122b-a10b","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":2.08,"openrouter_id":"qwen/qwen3.5-122b-a10b","context_length":262144}},{"model_name":"qwen3.5-27b","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":2.6,"openrouter_id":"qwen/qwen3.5-27b","context_length":262144}},{"model_name":"qwen3.5-plus-02-15","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...","quota_type":0,"model_ratio":0.13,"model_price":0,"owner_by":"qwen","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.26,"output_usd_per_m":1.56,"openrouter_id":"qwen/qwen3.5-plus-02-15","context_length":1000000}},{"model_name":"deepseek-v3.1-terminus","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...","quota_type":0,"model_ratio":0.135,"model_price":0,"owner_by":"deepseek","completion_ratio":3.7037,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.27,"output_usd_per_m":1,"openrouter_id":"deepseek/deepseek-v3.1-terminus","context_length":163840}},{"model_name":"deepseek-v3.2-exp","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...","quota_type":0,"model_ratio":0.135,"model_price":0,"owner_by":"deepseek","completion_ratio":1.5185,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.27,"output_usd_per_m":0.41,"openrouter_id":"deepseek/deepseek-v3.2-exp","context_length":163840}},{"model_name":"deepseek/deepseek-v3.1-terminus","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...","quota_type":0,"model_ratio":0.135,"model_price":0,"owner_by":"deepseek","completion_ratio":3.7037,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.27,"output_usd_per_m":1,"openrouter_id":"deepseek/deepseek-v3.1-terminus","context_length":163840}},{"model_name":"deepseek/deepseek-v3.2-exp","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...","quota_type":0,"model_ratio":0.135,"model_price":0,"owner_by":"deepseek","completion_ratio":1.5185,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.27,"output_usd_per_m":0.41,"openrouter_id":"deepseek/deepseek-v3.2-exp","context_length":163840}},{"model_name":"minimax-m2.5","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...","quota_type":0,"model_ratio":0.135,"model_price":0,"owner_by":"minimax","completion_ratio":3.5185,"cache_ratio":0.111111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.27,"output_usd_per_m":0.95,"openrouter_id":"minimax/minimax-m2.5","context_length":204800}},{"model_name":"minimax/minimax-m2.5","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...","quota_type":0,"model_ratio":0.135,"model_price":0,"owner_by":"minimax","completion_ratio":3.5185,"cache_ratio":0.111111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.27,"output_usd_per_m":0.95,"openrouter_id":"minimax/minimax-m2.5","context_length":204800}},{"model_name":"deepseek-chat","description":"高性价比编码","quota_type":0,"model_ratio":0.138889,"model_price":0,"owner_by":"deepseek","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,deepseek","grokcode":{"input_usd_per_m":0.27777778,"output_usd_per_m":1.11111111,"context_length":null}},{"model_name":"deepseek-chat-v3-0324","description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) model and performs really well...","quota_type":0,"model_ratio":0.145,"model_price":0,"owner_by":"deepseek","completion_ratio":3.931,"cache_ratio":0.37931,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.29,"output_usd_per_m":1.1400000000000001,"openrouter_id":"deepseek/deepseek-chat-v3-0324","context_length":163840}},{"model_name":"deepseek/deepseek-chat-v3-0324","description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) model and performs really well...","quota_type":0,"model_ratio":0.145,"model_price":0,"owner_by":"deepseek","completion_ratio":3.931,"cache_ratio":0.37931,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.29,"output_usd_per_m":1.1400000000000001,"openrouter_id":"deepseek/deepseek-chat-v3-0324","context_length":163840}},{"model_name":"qwen/qwen3-vl-30b-a3b-thinking","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...","quota_type":0,"model_ratio":0.145,"model_price":0,"owner_by":"qwen","completion_ratio":3.4483,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.29,"output_usd_per_m":1,"openrouter_id":"qwen/qwen3-vl-30b-a3b-thinking","context_length":262144}},{"model_name":"qwen3-vl-30b-a3b-thinking","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...","quota_type":0,"model_ratio":0.145,"model_price":0,"owner_by":"qwen","completion_ratio":3.4483,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.29,"output_usd_per_m":1,"openrouter_id":"qwen/qwen3-vl-30b-a3b-thinking","context_length":262144}},{"model_name":"amazon/nova-2-lite-v1","description":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text. Nova 2 Lite demonstrates standout capabilities in processing...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"amazon","completion_ratio":8.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"amazon/nova-2-lite-v1","context_length":1000000}},{"model_name":"beam-1.0-501b-a23b","description":"Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per token, and a...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"reflection","completion_ratio":3.3333,"cache_ratio":0.133333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,reflection","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1,"openrouter_id":"reflection/beam-1.0-501b-a23b","context_length":1048576}},{"model_name":"codestral-2508","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)","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":0.8999999999999999,"openrouter_id":"mistralai/codestral-2508","context_length":256000}},{"model_name":"cohere/command-a-plus","description":"Command A+ is Cohere's flagship model for enterprise agentic workflows. It accepts text and image inputs with a 192K context window, supports native tool calling with strict tool schemas, structured...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"cohere","completion_ratio":5,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.5,"openrouter_id":"cohere/command-a-plus","context_length":192000}},{"model_name":"command-a-plus","description":"Command A+ is Cohere's flagship model for enterprise agentic workflows. It accepts text and image inputs with a 192K context window, supports native tool calling with strict tool schemas, structured...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"cohere","completion_ratio":5,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.5,"openrouter_id":"cohere/command-a-plus","context_length":192000}},{"model_name":"cydonia-24b-v4.1","description":"Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"thedrummer","completion_ratio":1.6667,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thedrummer","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":0.5,"openrouter_id":"thedrummer/cydonia-24b-v4.1","context_length":131072}},{"model_name":"gemini-2.5-flash","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...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"google","completion_ratio":8.3333,"cache_ratio":0.1,"create_cache_ratio":0.277778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"google/gemini-2.5-flash","context_length":1048576}},{"model_name":"gemini-2.5-flash-image","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,...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"google","completion_ratio":8.3333,"cache_ratio":0.1,"create_cache_ratio":0.277778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"google/gemini-2.5-flash-image","context_length":32768}},{"model_name":"gemini-3.5-flash-lite","description":"Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"google","completion_ratio":8.3333,"cache_ratio":0.1,"create_cache_ratio":0.277778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"google/gemini-3.5-flash-lite","context_length":1048576}},{"model_name":"glm-4.6v","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...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"glm","completion_ratio":3,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":0.8999999999999999,"openrouter_id":"z-ai/glm-4.6v","context_length":131072}},{"model_name":"google/gemini-2.5-flash","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...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"google","completion_ratio":8.3333,"cache_ratio":0.1,"create_cache_ratio":0.277778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"google/gemini-2.5-flash","context_length":1048576}},{"model_name":"google/gemini-2.5-flash-image","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,...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"google","completion_ratio":8.3333,"cache_ratio":0.1,"create_cache_ratio":0.277778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"google/gemini-2.5-flash-image","context_length":32768}},{"model_name":"google/gemini-3.5-flash-lite","description":"Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"google","completion_ratio":8.3333,"cache_ratio":0.1,"create_cache_ratio":0.277778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"google/gemini-3.5-flash-lite","context_length":1048576}},{"model_name":"longcat-2.0","description":"LongCat 2.0 is a sparse mixture-of-experts language model from Meituan, with 48B active parameters out of 1.6T total. It is suited for coding, repository-level changes, long-horizon problem solving, and agentic...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"meituan","completion_ratio":4,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,meituan","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"openrouter_id":"meituan/longcat-2.0","context_length":1048756}},{"model_name":"meituan/longcat-2.0","description":"LongCat 2.0 is a sparse mixture-of-experts language model from Meituan, with 48B active parameters out of 1.6T total. It is suited for coding, repository-level changes, long-horizon problem solving, and agentic...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"meituan","completion_ratio":4,"cache_ratio":0.02,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,meituan","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"openrouter_id":"meituan/longcat-2.0","context_length":1048756}},{"model_name":"meta/muse-glimmer-30b","description":"Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"llama","completion_ratio":4,"cache_ratio":0.133333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"openrouter_id":"meta/muse-glimmer-30b","context_length":131072}},{"model_name":"MiniMax M2.5","description":"MiniMax M2.5 是低成本编码与复杂任务模型，仍被多个公开套餐收录。","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"MiniMax","completion_ratio":4,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,MiniMax","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"context_length":null}},{"model_name":"MiniMax M2.7","description":"MiniMax M2.7 是官方 API 与 Token Plan 的主力模型，适合低成本编码与 Agent。","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"MiniMax","completion_ratio":4,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,MiniMax","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"context_length":null}},{"model_name":"minimax-m3","description":"MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"minimax","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"openrouter_id":"minimax/minimax-m3","context_length":1048576}},{"model_name":"minimax/minimax-m3","description":"MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"minimax","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"openrouter_id":"minimax/minimax-m3","context_length":1048576}},{"model_name":"mistralai/codestral-2508","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)","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":0.8999999999999999,"openrouter_id":"mistralai/codestral-2508","context_length":256000}},{"model_name":"moondream/moondream3-preview","description":"Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over images, with...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"moondream","completion_ratio":8.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,moondream","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"moondream/moondream3-preview","context_length":32000}},{"model_name":"moondream/moondream3.1-9b-a2b","description":"Moondream 3.1 is a vision-language model from Moondream, the successor to Moondream 3 Preview, with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"moondream","completion_ratio":8.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,moondream","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"moondream/moondream3.1-9b-a2b","context_length":32000}},{"model_name":"moondream3-preview","description":"Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over images, with...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"moondream","completion_ratio":8.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,moondream","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"moondream/moondream3-preview","context_length":32000}},{"model_name":"moondream3.1-9b-a2b","description":"Moondream 3.1 is a vision-language model from Moondream, the successor to Moondream 3 Preview, with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"moondream","completion_ratio":8.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,moondream","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"moondream/moondream3.1-9b-a2b","context_length":32000}},{"model_name":"muse-glimmer-30b","description":"Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"llama","completion_ratio":4,"cache_ratio":0.133333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.2,"openrouter_id":"meta/muse-glimmer-30b","context_length":131072}},{"model_name":"Nano Banana","description":"Google Gemini 2.5 Flash Image 图片生成模型，适合低成本图片生成和编辑任务。","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"图片生成","completion_ratio":0.13,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,图片生成","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":0.039,"context_length":null}},{"model_name":"nemotron-3-super-120b-a12b","description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"nvidia","completion_ratio":8,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.4,"openrouter_id":"nvidia/nemotron-3-super-120b-a12b","context_length":1048576}},{"model_name":"nova-2-lite-v1","description":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text. Nova 2 Lite demonstrates standout capabilities in processing...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"amazon","completion_ratio":8.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.5,"openrouter_id":"amazon/nova-2-lite-v1","context_length":1000000}},{"model_name":"nvidia/nemotron-3-super-120b-a12b","description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"nvidia","completion_ratio":8,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":2.4,"openrouter_id":"nvidia/nemotron-3-super-120b-a12b","context_length":1048576}},{"model_name":"qwen/qwen3-coder","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...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"qwen","completion_ratio":3.3333,"cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1,"openrouter_id":"qwen/qwen3-coder","context_length":262144}},{"model_name":"qwen/qwen3.5-plus-20260420","description":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba. It accepts text, image, and video input and produces text output, with a 1M token context window. This...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"qwen","completion_ratio":6,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.7999999999999998,"openrouter_id":"qwen/qwen3.5-plus-20260420","context_length":1000000}},{"model_name":"qwen3-coder","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...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"qwen","completion_ratio":3.3333,"cache_ratio":0.333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1,"openrouter_id":"qwen/qwen3-coder","context_length":262144}},{"model_name":"qwen3.5-plus-20260420","description":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba. It accepts text, image, and video input and produces text output, with a 1M token context window. This...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"qwen","completion_ratio":6,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1.7999999999999998,"openrouter_id":"qwen/qwen3.5-plus-20260420","context_length":1000000}},{"model_name":"reflection/beam-1.0-501b-a23b","description":"Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per token, and a...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"reflection","completion_ratio":3.3333,"cache_ratio":0.133333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,reflection","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":1,"openrouter_id":"reflection/beam-1.0-501b-a23b","context_length":1048576}},{"model_name":"thedrummer/cydonia-24b-v4.1","description":"Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"thedrummer","completion_ratio":1.6667,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thedrummer","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":0.5,"openrouter_id":"thedrummer/cydonia-24b-v4.1","context_length":131072}},{"model_name":"z-ai/glm-4.6v","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...","quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"glm","completion_ratio":3,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.3,"output_usd_per_m":0.8999999999999999,"openrouter_id":"z-ai/glm-4.6v","context_length":131072}},{"model_name":"qwen/qwen3.7-plus","description":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its...","quota_type":0,"model_ratio":0.16,"model_price":0,"owner_by":"qwen","completion_ratio":4,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.32,"output_usd_per_m":1.28,"openrouter_id":"qwen/qwen3.7-plus","context_length":1000000}},{"model_name":"qwen3.7-plus","description":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its...","quota_type":0,"model_ratio":0.16,"model_price":0,"owner_by":"qwen","completion_ratio":4,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.32,"output_usd_per_m":1.28,"openrouter_id":"qwen/qwen3.7-plus","context_length":1000000}},{"model_name":"qwen/qwen3.6-plus","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...","quota_type":0,"model_ratio":0.1625,"model_price":0,"owner_by":"qwen","completion_ratio":6,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.325,"output_usd_per_m":1.95,"openrouter_id":"qwen/qwen3.6-plus","context_length":1000000}},{"model_name":"qwen3.6-plus","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...","quota_type":0,"model_ratio":0.1625,"model_price":0,"owner_by":"qwen","completion_ratio":6,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.325,"output_usd_per_m":1.95,"openrouter_id":"qwen/qwen3.6-plus","context_length":1000000}},{"model_name":"lfm-2.5-2.6b","description":"LFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or...","quota_type":0,"model_ratio":0.175,"model_price":0,"owner_by":"liquid","completion_ratio":1.4286,"cache_ratio":0.114286,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,liquid","grokcode":{"input_usd_per_m":0.35,"output_usd_per_m":0.5,"openrouter_id":"liquid/lfm-2.5-2.6b","context_length":128000}},{"model_name":"liquid/lfm-2.5-2.6b","description":"LFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or...","quota_type":0,"model_ratio":0.175,"model_price":0,"owner_by":"liquid","completion_ratio":1.4286,"cache_ratio":0.114286,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,liquid","grokcode":{"input_usd_per_m":0.35,"output_usd_per_m":0.5,"openrouter_id":"liquid/lfm-2.5-2.6b","context_length":128000}},{"model_name":"llama-4-maverick","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...","quota_type":0,"model_ratio":0.175,"model_price":0,"owner_by":"llama","completion_ratio":2.8571,"cache_ratio":0.485714,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.35,"output_usd_per_m":1,"openrouter_id":"meta-llama/llama-4-maverick","context_length":524288}},{"model_name":"meta-llama/llama-4-maverick","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...","quota_type":0,"model_ratio":0.175,"model_price":0,"owner_by":"llama","completion_ratio":2.8571,"cache_ratio":0.485714,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.35,"output_usd_per_m":1,"openrouter_id":"meta-llama/llama-4-maverick","context_length":524288}},{"model_name":"remm-slerp-l2-13b","description":"A recreation trial of the original MythoMax-L2-B13 but with updated models. #merge","quota_type":0,"model_ratio":0.175,"model_price":0,"owner_by":"undi95","completion_ratio":1.8571,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,undi95","grokcode":{"input_usd_per_m":0.35,"output_usd_per_m":0.65,"openrouter_id":"undi95/remm-slerp-l2-13b","context_length":6144}},{"model_name":"undi95/remm-slerp-l2-13b","description":"A recreation trial of the original MythoMax-L2-B13 but with updated models. #merge","quota_type":0,"model_ratio":0.175,"model_price":0,"owner_by":"undi95","completion_ratio":1.8571,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,undi95","grokcode":{"input_usd_per_m":0.35,"output_usd_per_m":0.65,"openrouter_id":"undi95/remm-slerp-l2-13b","context_length":6144}},{"model_name":"mistral-small-3.1-24b-instruct","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...","quota_type":0,"model_ratio":0.1755,"model_price":0,"owner_by":"mistral","completion_ratio":1.5812,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.351,"output_usd_per_m":0.5549999999999999,"openrouter_id":"mistralai/mistral-small-3.1-24b-instruct","context_length":128000}},{"model_name":"mistralai/mistral-small-3.1-24b-instruct","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...","quota_type":0,"model_ratio":0.1755,"model_price":0,"owner_by":"mistral","completion_ratio":1.5812,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.351,"output_usd_per_m":0.5549999999999999,"openrouter_id":"mistralai/mistral-small-3.1-24b-instruct","context_length":128000}},{"model_name":"qwen-2.5-72b-instruct","description":"Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...","quota_type":0,"model_ratio":0.18,"model_price":0,"owner_by":"qwen","completion_ratio":1.1111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.36,"output_usd_per_m":0.39999999999999997,"openrouter_id":"qwen/qwen-2.5-72b-instruct","context_length":32768}},{"model_name":"qwen/qwen-2.5-72b-instruct","description":"Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...","quota_type":0,"model_ratio":0.18,"model_price":0,"owner_by":"qwen","completion_ratio":1.1111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.36,"output_usd_per_m":0.39999999999999997,"openrouter_id":"qwen/qwen-2.5-72b-instruct","context_length":32768}},{"model_name":"glm-5.3-flashx","description":"GLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s. Built on the same hybrid sparse and linear attention architecture...","quota_type":0,"model_ratio":0.185,"model_price":0,"owner_by":"glm","completion_ratio":3.3784,"cache_ratio":0.202703,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.37,"output_usd_per_m":1.25,"openrouter_id":"z-ai/glm-5.3-flashx","context_length":1048576}},{"model_name":"glm-5.3-flashx-legal","description":"GLM-5.3-FlashX-Legal (GLM-Legal) is Z.ai's legal-domain variant of GLM-5.3-FlashX, a native multimodal model that accepts text, image, and video input and produces text. It adds legal-domain mid-training and post-training with reinforcement...","quota_type":0,"model_ratio":0.185,"model_price":0,"owner_by":"glm","completion_ratio":3.3784,"cache_ratio":0.202703,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.37,"output_usd_per_m":1.25,"openrouter_id":"z-ai/glm-5.3-flashx-legal","context_length":1048576}},{"model_name":"z-ai/glm-5.3-flashx","description":"GLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s. Built on the same hybrid sparse and linear attention architecture...","quota_type":0,"model_ratio":0.185,"model_price":0,"owner_by":"glm","completion_ratio":3.3784,"cache_ratio":0.202703,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.37,"output_usd_per_m":1.25,"openrouter_id":"z-ai/glm-5.3-flashx","context_length":1048576}},{"model_name":"z-ai/glm-5.3-flashx-legal","description":"GLM-5.3-FlashX-Legal (GLM-Legal) is Z.ai's legal-domain variant of GLM-5.3-FlashX, a native multimodal model that accepts text, image, and video input and produces text. It adds legal-domain mid-training and post-training with reinforcement...","quota_type":0,"model_ratio":0.185,"model_price":0,"owner_by":"glm","completion_ratio":3.3784,"cache_ratio":0.202703,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.37,"output_usd_per_m":1.25,"openrouter_id":"z-ai/glm-5.3-flashx-legal","context_length":1048576}},{"model_name":"google/gemini-3.6-flash:batch","description":"Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development. It is designed to produce polished outputs with fewer unnecessary edits and...","quota_type":0,"model_ratio":0.1875,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.111111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.375,"output_usd_per_m":1.875,"openrouter_id":"google/gemini-3.6-flash:batch","context_length":1048576}},{"model_name":"google/gemini-3.7-flash:batch","description":"Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...","quota_type":0,"model_ratio":0.1875,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.111111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.375,"output_usd_per_m":1.875,"openrouter_id":"google/gemini-3.7-flash:batch","context_length":1048576}},{"model_name":"google/gemini-3.8-flash:batch","description":"Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.","quota_type":0,"model_ratio":0.1875,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.111111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.375,"output_usd_per_m":1.875,"openrouter_id":"google/gemini-3.8-flash:batch","context_length":1048576}},{"model_name":"openai/gpt-5.4-mini:batch","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,...","quota_type":0,"model_ratio":0.1875,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.375,"output_usd_per_m":2.25,"openrouter_id":"openai/gpt-5.4-mini:batch","context_length":400000}},{"model_name":"devstral-2512","description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mistral","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"mistralai/devstral-2512","context_length":262144}},{"model_name":"mancer/weaver","description":"An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory. Meant for use in roleplay/narrative situations.","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mancer","completion_ratio":1.875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mancer","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":0.75,"openrouter_id":"mancer/weaver","context_length":8000}},{"model_name":"mimo-v2.5","description":"MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"xiaomi","completion_ratio":5,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"xiaomi/mimo-v2.5","context_length":1000000}},{"model_name":"mistral-medium-3","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...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mistral","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"mistralai/mistral-medium-3","context_length":131072}},{"model_name":"mistral-medium-3.1","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...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mistral","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"mistralai/mistral-medium-3.1","context_length":131072}},{"model_name":"mistralai/devstral-2512","description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mistral","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"mistralai/devstral-2512","context_length":262144}},{"model_name":"mistralai/mistral-medium-3","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...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mistral","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"mistralai/mistral-medium-3","context_length":131072}},{"model_name":"mistralai/mistral-medium-3.1","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...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mistral","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"mistralai/mistral-medium-3.1","context_length":131072}},{"model_name":"openai/gpt-4.1-mini","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...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":1.5999999999999999,"openrouter_id":"openai/gpt-4.1-mini","context_length":1047576}},{"model_name":"thedrummer/unslopnemo-12b","description":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"thedrummer","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thedrummer","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":0.39999999999999997,"openrouter_id":"thedrummer/unslopnemo-12b","context_length":1024000}},{"model_name":"unslopnemo-12b","description":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"thedrummer","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thedrummer","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":0.39999999999999997,"openrouter_id":"thedrummer/unslopnemo-12b","context_length":1024000}},{"model_name":"weaver","description":"An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory. Meant for use in roleplay/narrative situations.","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"mancer","completion_ratio":1.875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mancer","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":0.75,"openrouter_id":"mancer/weaver","context_length":8000}},{"model_name":"xiaomi/mimo-v2.5","description":"MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"xiaomi","completion_ratio":5,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.39999999999999997,"output_usd_per_m":2,"openrouter_id":"xiaomi/mimo-v2.5","context_length":1000000}},{"model_name":"gpt-4.1-mini","description":"gpt-4.1-mini","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":0.4,"output_usd_per_m":1.6,"context_length":null}},{"model_name":"Qwen3.7 Plus","description":"Qwen 的均衡版本，适合成本敏感的 Coding Agent 和日常中文开发任务，已补齐阿里云百炼官方按量价格。","quota_type":0,"model_ratio":0.2,"model_price":0,"owner_by":"Qwen","completion_ratio":4,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Qwen","grokcode":{"input_usd_per_m":0.4,"output_usd_per_m":1.6,"context_length":null}},{"model_name":"hy4","description":"The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity...","quota_type":0,"model_ratio":0.2085,"model_price":0,"owner_by":"tencent","completion_ratio":2.9988,"cache_ratio":0.05036,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.417,"output_usd_per_m":1.2505,"openrouter_id":"tencent/hy4","context_length":1048576}},{"model_name":"tencent/hy4","description":"The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity...","quota_type":0,"model_ratio":0.2085,"model_price":0,"owner_by":"tencent","completion_ratio":2.9988,"cache_ratio":0.05036,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.417,"output_usd_per_m":1.2505,"openrouter_id":"tencent/hy4","context_length":1048576}},{"model_name":"MiMo V2.5 Pro","description":"Xiaomi MiMo V2.5 系列旗舰基座模型，官方介绍为 1T 总参数、42B 激活与 1M 上下文，适合复杂 Agent、编码和长上下文任务。","quota_type":0,"model_ratio":0.2175,"model_price":0,"owner_by":"MiMo","completion_ratio":2,"cache_ratio":0.008276,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,MiMo","grokcode":{"input_usd_per_m":0.435,"output_usd_per_m":0.87,"context_length":null}},{"model_name":"mimo-v2.6-pro","description":"MiMo-V2.6-Pro is the flagship foundation model developed by Xiaomi. Built at a scale of over 1T parameters, it is designed to push the ceiling of capability for the most demanding...","quota_type":0,"model_ratio":0.2175,"model_price":0,"owner_by":"xiaomi","completion_ratio":2,"cache_ratio":0.008276,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.435,"output_usd_per_m":0.87,"openrouter_id":"xiaomi/mimo-v2.6-pro","context_length":1050000}},{"model_name":"xiaomi/mimo-v2.6-pro","description":"MiMo-V2.6-Pro is the flagship foundation model developed by Xiaomi. Built at a scale of over 1T parameters, it is designed to push the ceiling of capability for the most demanding...","quota_type":0,"model_ratio":0.2175,"model_price":0,"owner_by":"xiaomi","completion_ratio":2,"cache_ratio":0.008276,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.435,"output_usd_per_m":0.87,"openrouter_id":"xiaomi/mimo-v2.6-pro","context_length":1050000}},{"model_name":"inkling-small","description":"Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...","quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"thinkingmachines","completion_ratio":2.6667,"cache_ratio":0.222222,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thinkingmachines","grokcode":{"input_usd_per_m":0.44999999999999996,"output_usd_per_m":1.2,"openrouter_id":"thinkingmachines/inkling-small","context_length":524288}},{"model_name":"qwen/qwen3-235b-a22b-thinking-2507","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...","quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"qwen","completion_ratio":7.7778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.44999999999999996,"output_usd_per_m":3.5,"openrouter_id":"qwen/qwen3-235b-a22b-thinking-2507","context_length":128000}},{"model_name":"qwen/qwen3.6-27b","description":"Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026. It features hybrid multimodal capabilities — accepting text, image, and video inputs...","quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"qwen","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.44999999999999996,"output_usd_per_m":2.7,"openrouter_id":"qwen/qwen3.6-27b","context_length":262144}},{"model_name":"qwen3-235b-a22b-thinking-2507","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...","quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"qwen","completion_ratio":7.7778,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.44999999999999996,"output_usd_per_m":3.5,"openrouter_id":"qwen/qwen3-235b-a22b-thinking-2507","context_length":128000}},{"model_name":"qwen3.6-27b","description":"Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026. It features hybrid multimodal capabilities — accepting text, image, and video inputs...","quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"qwen","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.44999999999999996,"output_usd_per_m":2.7,"openrouter_id":"qwen/qwen3.6-27b","context_length":262144}},{"model_name":"thinkingmachines/inkling-small","description":"Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...","quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"thinkingmachines","completion_ratio":2.6667,"cache_ratio":0.222222,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thinkingmachines","grokcode":{"input_usd_per_m":0.44999999999999996,"output_usd_per_m":1.2,"openrouter_id":"thinkingmachines/inkling-small","context_length":524288}},{"model_name":"z-ai/glm-5.3:batch","description":"GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...","quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"glm","completion_ratio":4.4444,"cache_ratio":0.222222,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.44999999999999996,"output_usd_per_m":2,"openrouter_id":"z-ai/glm-5.3:batch","context_length":1048576}},{"model_name":"mimo-v2.5-pro","description":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as ClawEval, GDPVal, and SWE-bench Pro....","quota_type":0,"model_ratio":0.24,"model_price":0,"owner_by":"xiaomi","completion_ratio":3.75,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.48,"output_usd_per_m":1.7999999999999998,"openrouter_id":"xiaomi/mimo-v2.5-pro","context_length":1050000}},{"model_name":"xiaomi/mimo-v2.5-pro","description":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as ClawEval, GDPVal, and SWE-bench Pro....","quota_type":0,"model_ratio":0.24,"model_price":0,"owner_by":"xiaomi","completion_ratio":3.75,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":0.48,"output_usd_per_m":1.7999999999999998,"openrouter_id":"xiaomi/mimo-v2.5-pro","context_length":1050000}},{"model_name":"anthropic/claude-haiku-4.5:batch","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.5,"openrouter_id":"anthropic/claude-haiku-4.5:batch","context_length":200000}},{"model_name":"bytedance-seed/seed-2-1-turbo","description":"Seed 2.1 Turbo is a multimodal model from ByteDance Seed for coding and long-horizon agent workflows. It is suited for end-to-end software delivery, multi-step task execution, and understanding visual and...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.5,"openrouter_id":"bytedance-seed/seed-2-1-turbo","context_length":262144}},{"model_name":"bytedance-seed/seed-2.0-code","description":"Seed 2.0 Code is a model from ByteDance Seed optimized for agentic coding. It is suited for frontend development, multilingual programming tasks, and coding-agent workflows in tools such as Claude...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"bytedance-seed/seed-2.0-code","context_length":262144}},{"model_name":"deepseek-r1-0528","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"deepseek","completion_ratio":4.3,"cache_ratio":0.7,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.1500000000000004,"openrouter_id":"deepseek/deepseek-r1-0528","context_length":163840}},{"model_name":"deepseek/deepseek-r1-0528","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"deepseek","completion_ratio":4.3,"cache_ratio":0.7,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.1500000000000004,"openrouter_id":"deepseek/deepseek-r1-0528","context_length":163840}},{"model_name":"Gemini 3 Flash Preview","description":"Google Gemini Flash 预览模型，适合速度优先的多模态和日常 Agent 场景，价格以官方预览页为准。","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"Gemini","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,Gemini","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"context_length":null}},{"model_name":"gemini-3-flash-preview","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.166667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"google/gemini-3-flash-preview","context_length":1048576}},{"model_name":"gemini-3.1-flash-image","description":"Gemini 3.1 Flash Image, a.k.a. \"Nano Banana 2,\" is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines advanced...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"google/gemini-3.1-flash-image","context_length":131072}},{"model_name":"gemini-3.1-flash-image-preview","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"google/gemini-3.1-flash-image-preview","context_length":65536}},{"model_name":"glm-4.6","description":"Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"glm","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2,"openrouter_id":"z-ai/glm-4.6","context_length":204800}},{"model_name":"google/gemini-3-flash-preview","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.166667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"google/gemini-3-flash-preview","context_length":1048576}},{"model_name":"google/gemini-3.1-flash-image","description":"Gemini 3.1 Flash Image, a.k.a. \"Nano Banana 2,\" is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines advanced...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"google/gemini-3.1-flash-image","context_length":131072}},{"model_name":"google/gemini-3.1-flash-image-preview","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"google/gemini-3.1-flash-image-preview","context_length":65536}},{"model_name":"kimi-k2.5","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"kimi","completion_ratio":5,"cache_ratio":0.3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.5,"openrouter_id":"moonshotai/kimi-k2.5","context_length":262144}},{"model_name":"mistral-large-2512","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.","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":1.5,"openrouter_id":"mistralai/mistral-large-2512","context_length":262144}},{"model_name":"mistralai/mistral-large-2512","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.","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":1.5,"openrouter_id":"mistralai/mistral-large-2512","context_length":262144}},{"model_name":"moonshotai/kimi-k2.5","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"kimi","completion_ratio":5,"cache_ratio":0.3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.5,"openrouter_id":"moonshotai/kimi-k2.5","context_length":262144}},{"model_name":"Nano Banana 2","description":"Google Gemini 3.1 Flash Image 图片生成模型，适合较低成本的图像生成、编辑和批量图片任务。","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"图片生成","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,图片生成","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":0,"context_length":null}},{"model_name":"openai/gpt-6.1-terra-pro:batch","description":"GPT-6.1 Terra Pro is the same underlying model as [GPT-6.1 Terra](https://openrouter.ai/openai/gpt-6.1-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.5,"openrouter_id":"openai/gpt-6.1-terra-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-6.1-terra:batch","description":"GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.5,"openrouter_id":"openai/gpt-6.1-terra:batch","context_length":1050000}},{"model_name":"qwen/qwen3.5-397b-a17b","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"qwen","completion_ratio":7.2,"cache_ratio":0.6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3.5999999999999996,"openrouter_id":"qwen/qwen3.5-397b-a17b","context_length":262144}},{"model_name":"qwen3.5-397b-a17b","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...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"qwen","completion_ratio":7.2,"cache_ratio":0.6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3.5999999999999996,"openrouter_id":"qwen/qwen3.5-397b-a17b","context_length":262144}},{"model_name":"Qwen3.6 Plus","description":"Qwen Plus 系列模型，适合中低成本编码和中文模型调用，已补齐阿里云百炼官方按量价格。","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"Qwen","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Qwen","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"context_length":null}},{"model_name":"seed-2-1-turbo","description":"Seed 2.1 Turbo is a multimodal model from ByteDance Seed for coding and long-horizon agent workflows. It is suited for end-to-end software delivery, multi-step task execution, and understanding visual and...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2.5,"openrouter_id":"bytedance-seed/seed-2-1-turbo","context_length":262144}},{"model_name":"seed-2.0-code","description":"Seed 2.0 Code is a model from ByteDance Seed optimized for agentic coding. It is suited for frontend development, multilingual programming tasks, and coding-agent workflows in tools such as Claude...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"bytedance-seed","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,bytedance-seed","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":3,"openrouter_id":"bytedance-seed/seed-2.0-code","context_length":262144}},{"model_name":"z-ai/glm-4.6","description":"Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...","quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"glm","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.5,"output_usd_per_m":2,"openrouter_id":"z-ai/glm-4.6","context_length":204800}},{"model_name":"GPT Image 1","description":"GPT Image 1","quota_type":0,"model_ratio":0.273398,"model_price":0,"owner_by":"openai","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":0.54679688,"output_usd_per_m":0,"context_length":null}},{"model_name":"minimax-m1","description":"MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference. It leverages a hybrid Mixture-of-Experts (MoE) architecture paired with a custom \"lightning attention\" mechanism, allowing it...","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"minimax","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":2.2,"openrouter_id":"minimax/minimax-m1","context_length":1000000}},{"model_name":"minimax/minimax-m1","description":"MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference. It leverages a hybrid Mixture-of-Experts (MoE) architecture paired with a custom \"lightning attention\" mechanism, allowing it...","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"minimax","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,minimax","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":2.2,"openrouter_id":"minimax/minimax-m1","context_length":1000000}},{"model_name":"o3-mini","description":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding. This model supports the `reasoning_effort` parameter, which can be set to...","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":2.2,"openrouter_id":"openai/o3-mini:batch","context_length":200000}},{"model_name":"o4-mini","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...","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":2.2,"openrouter_id":"openai/o4-mini:batch","context_length":200000}},{"model_name":"openai/o3-mini:batch","description":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding. This model supports the `reasoning_effort` parameter, which can be set to...","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":2.2,"openrouter_id":"openai/o3-mini:batch","context_length":200000}},{"model_name":"openai/o4-mini:batch","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...","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":2.2,"openrouter_id":"openai/o4-mini:batch","context_length":200000}},{"model_name":"skyfall-36b-v2","description":"Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"thedrummer","completion_ratio":1.4545,"cache_ratio":0.454545,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thedrummer","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":0.7999999999999999,"openrouter_id":"thedrummer/skyfall-36b-v2","context_length":32768}},{"model_name":"thedrummer/skyfall-36b-v2","description":"Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.","quota_type":0,"model_ratio":0.275,"model_price":0,"owner_by":"thedrummer","completion_ratio":1.4545,"cache_ratio":0.454545,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thedrummer","grokcode":{"input_usd_per_m":0.55,"output_usd_per_m":0.7999999999999999,"openrouter_id":"thedrummer/skyfall-36b-v2","context_length":32768}},{"model_name":"Qwen3 Coder Plus","description":"Qwen Coder 系列模型，适合代码生成和 Agent 任务，已补齐阿里云百炼官方按量价格。","quota_type":0,"model_ratio":0.277778,"model_price":0,"owner_by":"Qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Qwen","grokcode":{"input_usd_per_m":0.55555556,"output_usd_per_m":2.22222222,"context_length":null}},{"model_name":"kimi-k2","description":"Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...","quota_type":0,"model_ratio":0.285,"model_price":0,"owner_by":"kimi","completion_ratio":4.0351,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.5700000000000001,"output_usd_per_m":2.3,"openrouter_id":"moonshotai/kimi-k2","context_length":131072}},{"model_name":"moonshotai/kimi-k2","description":"Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...","quota_type":0,"model_ratio":0.285,"model_price":0,"owner_by":"kimi","completion_ratio":4.0351,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.5700000000000001,"output_usd_per_m":2.3,"openrouter_id":"moonshotai/kimi-k2","context_length":131072}},{"model_name":"llama-3.3-70b-instruct","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...","quota_type":0,"model_ratio":0.295,"model_price":0,"owner_by":"llama","completion_ratio":1.339,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.59,"output_usd_per_m":0.7899999999999999,"openrouter_id":"meta-llama/llama-3.3-70b-instruct","context_length":131072}},{"model_name":"meta-llama/llama-3.3-70b-instruct","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...","quota_type":0,"model_ratio":0.295,"model_price":0,"owner_by":"llama","completion_ratio":1.339,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.59,"output_usd_per_m":0.7899999999999999,"openrouter_id":"meta-llama/llama-3.3-70b-instruct","context_length":131072}},{"model_name":"elorian/glimpse","description":"Glimpse is a multimodal visual reasoning model from Elorian, built for visual understanding and complex visual reasoning in the physical world. It reasons directly over images and video rather than...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"elorian","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,elorian","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":3,"openrouter_id":"elorian/glimpse","context_length":262144}},{"model_name":"glimpse","description":"Glimpse is a multimodal visual reasoning model from Elorian, built for visual understanding and complex visual reasoning in the physical world. It reasons directly over images and video rather than...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"elorian","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,elorian","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":3,"openrouter_id":"elorian/glimpse","context_length":262144}},{"model_name":"glm-4.5","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"glm","completion_ratio":3.6667,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.2,"openrouter_id":"z-ai/glm-4.5","context_length":131072}},{"model_name":"glm-4.5v","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,...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"glm","completion_ratio":3,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":1.7999999999999998,"openrouter_id":"z-ai/glm-4.5v","context_length":65536}},{"model_name":"glm-4.7","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"glm","completion_ratio":3.6667,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.2,"openrouter_id":"z-ai/glm-4.7","context_length":204800}},{"model_name":"gpt-audio-mini","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.4,"openrouter_id":"openai/gpt-audio-mini","context_length":128000}},{"model_name":"Kimi K2.5","description":"Kimi K2.5 多模态与 Agent 模型，仍被多个公开编码套餐收录。","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"Kimi","completion_ratio":5,"cache_ratio":0.166667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Kimi","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":3,"context_length":null}},{"model_name":"kimi-k2-0905","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"kimi","completion_ratio":4.1667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.5,"openrouter_id":"moonshotai/kimi-k2-0905","context_length":262144}},{"model_name":"kimi-k2-thinking","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"kimi","completion_ratio":4.1667,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.5,"openrouter_id":"moonshotai/kimi-k2-thinking","context_length":262144}},{"model_name":"kimi-k2.7-code","description":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"kimi","completion_ratio":5.8333,"cache_ratio":0.266667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":3.5,"openrouter_id":"moonshotai/kimi-k2.7-code","context_length":262144}},{"model_name":"moonshotai/kimi-k2-0905","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"kimi","completion_ratio":4.1667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.5,"openrouter_id":"moonshotai/kimi-k2-0905","context_length":262144}},{"model_name":"moonshotai/kimi-k2-thinking","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"kimi","completion_ratio":4.1667,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.5,"openrouter_id":"moonshotai/kimi-k2-thinking","context_length":262144}},{"model_name":"moonshotai/kimi-k2.7-code","description":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"kimi","completion_ratio":5.8333,"cache_ratio":0.266667,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":3.5,"openrouter_id":"moonshotai/kimi-k2.7-code","context_length":262144}},{"model_name":"nemotron-3-ultra-550b-a55b","description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"nvidia","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.4,"openrouter_id":"nvidia/nemotron-3-ultra-550b-a55b","context_length":262144}},{"model_name":"nvidia/nemotron-3-ultra-550b-a55b","description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"nvidia","completion_ratio":4,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.4,"openrouter_id":"nvidia/nemotron-3-ultra-550b-a55b","context_length":262144}},{"model_name":"openai/gpt-audio-mini","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.4,"openrouter_id":"openai/gpt-audio-mini","context_length":128000}},{"model_name":"palmyra-x5","description":"Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise. It delivers industry-leading speed and efficiency on context windows up to 1 million...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"writer","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,writer","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":6,"openrouter_id":"writer/palmyra-x5","context_length":1040000}},{"model_name":"writer/palmyra-x5","description":"Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise. It delivers industry-leading speed and efficiency on context windows up to 1 million...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"writer","completion_ratio":10,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,writer","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":6,"openrouter_id":"writer/palmyra-x5","context_length":1040000}},{"model_name":"z-ai/glm-4.5","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"glm","completion_ratio":3.6667,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.2,"openrouter_id":"z-ai/glm-4.5","context_length":131072}},{"model_name":"z-ai/glm-4.5v","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,...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"glm","completion_ratio":3,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":1.7999999999999998,"openrouter_id":"z-ai/glm-4.5v","context_length":65536}},{"model_name":"z-ai/glm-4.7","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...","quota_type":0,"model_ratio":0.3,"model_price":0,"owner_by":"glm","completion_ratio":3.6667,"cache_ratio":0.183333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":0.6,"output_usd_per_m":2.2,"openrouter_id":"z-ai/glm-4.7","context_length":204800}},{"model_name":"openai/gpt-5:batch","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...","quota_type":0,"model_ratio":0.3125,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.625,"output_usd_per_m":5,"openrouter_id":"openai/gpt-5:batch","context_length":400000}},{"model_name":"openai/gpt-5.1:batch","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...","quota_type":0,"model_ratio":0.3125,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.625,"output_usd_per_m":5,"openrouter_id":"openai/gpt-5.1:batch","context_length":400000}},{"model_name":"gemma-2-27b-it","description":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini). Gemma models are well-suited for a variety of...","quota_type":0,"model_ratio":0.325,"model_price":0,"owner_by":"gemma","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.65,"output_usd_per_m":0.65,"openrouter_id":"google/gemma-2-27b-it","context_length":8192}},{"model_name":"google/gemma-2-27b-it","description":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini). Gemma models are well-suited for a variety of...","quota_type":0,"model_ratio":0.325,"model_price":0,"owner_by":"gemma","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0.65,"output_usd_per_m":0.65,"openrouter_id":"google/gemma-2-27b-it","context_length":8192}},{"model_name":"kimi-k2.6","description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and...","quota_type":0,"model_ratio":0.325,"model_price":0,"owner_by":"kimi","completion_ratio":5.2462,"cache_ratio":0.230769,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.65,"output_usd_per_m":3.41,"openrouter_id":"moonshotai/kimi-k2.6","context_length":262144}},{"model_name":"moonshotai/kimi-k2.6","description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and...","quota_type":0,"model_ratio":0.325,"model_price":0,"owner_by":"kimi","completion_ratio":5.2462,"cache_ratio":0.230769,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":0.65,"output_usd_per_m":3.41,"openrouter_id":"moonshotai/kimi-k2.6","context_length":262144}},{"model_name":"qwen-2.5-coder-32b-instruct","description":"Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). Qwen2.5-Coder brings the following improvements upon CodeQwen1.5: - Significantly improvements in **code generation**, **code reasoning**...","quota_type":0,"model_ratio":0.33,"model_price":0,"owner_by":"qwen","completion_ratio":1.5152,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.66,"output_usd_per_m":1,"openrouter_id":"qwen/qwen-2.5-coder-32b-instruct","context_length":32768}},{"model_name":"qwen/qwen-2.5-coder-32b-instruct","description":"Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). Qwen2.5-Coder brings the following improvements upon CodeQwen1.5: - Significantly improvements in **code generation**, **code reasoning**...","quota_type":0,"model_ratio":0.33,"model_price":0,"owner_by":"qwen","completion_ratio":1.5152,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.66,"output_usd_per_m":1,"openrouter_id":"qwen/qwen-2.5-coder-32b-instruct","context_length":32768}},{"model_name":"aion-3.0-mini","description":"Aion-3.0 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the DeepSeek family of models. It uses a collaborative generation process in which multiple specialized models each...","quota_type":0,"model_ratio":0.35,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.257143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7,"output_usd_per_m":1.4,"openrouter_id":"aion-labs/aion-3.0-mini","context_length":131072}},{"model_name":"aion-3.5-mini","description":"Aion 3.5 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It is the smaller, lower-cost sibling of Aion 3.5 and uses...","quota_type":0,"model_ratio":0.35,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.257143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7,"output_usd_per_m":1.4,"openrouter_id":"aion-labs/aion-3.5-mini","context_length":262144}},{"model_name":"aion-labs/aion-3.0-mini","description":"Aion-3.0 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the DeepSeek family of models. It uses a collaborative generation process in which multiple specialized models each...","quota_type":0,"model_ratio":0.35,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.257143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7,"output_usd_per_m":1.4,"openrouter_id":"aion-labs/aion-3.0-mini","context_length":131072}},{"model_name":"aion-labs/aion-3.5-mini","description":"Aion 3.5 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It is the smaller, lower-cost sibling of Aion 3.5 and uses...","quota_type":0,"model_ratio":0.35,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.257143,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7,"output_usd_per_m":1.4,"openrouter_id":"aion-labs/aion-3.5-mini","context_length":262144}},{"model_name":"mistral-large-4-0","description":"Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads. It offers a 1M-token context window with up...","quota_type":0,"model_ratio":0.35,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.7,"output_usd_per_m":2.0999999999999996,"openrouter_id":"mistralai/mistral-large-4-0","context_length":1048576}},{"model_name":"mistralai/mistral-large-4-0","description":"Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads. It offers a 1M-token context window with up...","quota_type":0,"model_ratio":0.35,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.7,"output_usd_per_m":2.0999999999999996,"openrouter_id":"mistralai/mistral-large-4-0","context_length":1048576}},{"model_name":"llama-3.1-70b-instruct","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 70B instruct-tuned version is optimized for high quality dialogue usecases. It has demonstrated strong...","quota_type":0,"model_ratio":0.36,"model_price":0,"owner_by":"llama","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.72,"output_usd_per_m":0.72,"openrouter_id":"meta-llama/llama-3.1-70b-instruct","context_length":131072}},{"model_name":"meta-llama/llama-3.1-70b-instruct","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 70B instruct-tuned version is optimized for high quality dialogue usecases. It has demonstrated strong...","quota_type":0,"model_ratio":0.36,"model_price":0,"owner_by":"llama","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":0.72,"output_usd_per_m":0.72,"openrouter_id":"meta-llama/llama-3.1-70b-instruct","context_length":131072}},{"model_name":"~google/gemini-flash-latest","description":"This model always redirects to the latest model in the Gemini Flash family.","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"~google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~google","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"openrouter_id":"~google/gemini-flash-latest","context_length":1048576}},{"model_name":"~openai/gpt-mini-latest","description":"This model always redirects to the latest model in the GPT Mini family.","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"~openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":4.5,"openrouter_id":"~openai/gpt-mini-latest","context_length":400000}},{"model_name":"Gemini 3.6 Flash","description":"Google Gemini Flash 模型，适合低延迟、多模态和高吞吐任务。","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"Gemini","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Gemini","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"context_length":null}},{"model_name":"Gemini 3.7 Flash","description":"Google Gemini Flash 模型，适合低延迟、多模态和高吞吐任务。","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"Gemini","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Gemini","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"context_length":null}},{"model_name":"Gemini 3.8 Flash","description":"Google Gemini Flash 模型，适合低延迟、多模态和高吞吐任务。","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"Gemini","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Gemini","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"context_length":null}},{"model_name":"gemini-3.6-flash","description":"Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development. It is designed to produce polished outputs with fewer unnecessary edits and...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"openrouter_id":"google/gemini-3.6-flash","context_length":1048576}},{"model_name":"gemini-3.7-flash","description":"Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"openrouter_id":"google/gemini-3.7-flash","context_length":1048576}},{"model_name":"gemini-flash-latest","description":"This model always redirects to the latest model in the Gemini Flash family.","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"~google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~google","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"openrouter_id":"~google/gemini-flash-latest","context_length":1048576}},{"model_name":"google/gemini-3.5-flash:batch","description":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":4.5,"openrouter_id":"google/gemini-3.5-flash:batch","context_length":1048576}},{"model_name":"google/gemini-3.6-flash","description":"Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development. It is designed to produce polished outputs with fewer unnecessary edits and...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"openrouter_id":"google/gemini-3.6-flash","context_length":1048576}},{"model_name":"google/gemini-3.7-flash","description":"Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"openrouter_id":"google/gemini-3.7-flash","context_length":1048576}},{"model_name":"GPT-5.4 Mini","description":"OpenAI 成本更低的通用模型，适合批量抽取、轻量代码辅助、分类和日常自动化。","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"OpenAI","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,OpenAI","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":4.5,"context_length":null}},{"model_name":"gpt-5.4-mini","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,...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":4.5,"openrouter_id":"openai/gpt-5.4-mini","context_length":400000}},{"model_name":"gpt-mini-latest","description":"This model always redirects to the latest model in the GPT Mini family.","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"~openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":4.5,"openrouter_id":"~openai/gpt-mini-latest","context_length":400000}},{"model_name":"mistralai/mistral-medium-3-5:batch","description":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"mistral","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":3.75,"openrouter_id":"mistralai/mistral-medium-3-5:batch","context_length":262144}},{"model_name":"openai/gpt-5.4-mini","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,...","quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.75,"output_usd_per_m":4.5,"openrouter_id":"openai/gpt-5.4-mini","context_length":400000}},{"model_name":"hy4-preview","description":"Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks that...","quota_type":0,"model_ratio":0.3753,"model_price":0,"owner_by":"tencent","completion_ratio":2.9988,"cache_ratio":0.05036,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.7505999999999999,"output_usd_per_m":2.2509,"openrouter_id":"tencent/hy4-preview","context_length":1048576}},{"model_name":"tencent/hy4-preview","description":"Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks that...","quota_type":0,"model_ratio":0.3753,"model_price":0,"owner_by":"tencent","completion_ratio":2.9988,"cache_ratio":0.05036,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,tencent","grokcode":{"input_usd_per_m":0.7505999999999999,"output_usd_per_m":2.2509,"openrouter_id":"tencent/hy4-preview","context_length":1048576}},{"model_name":"~unbiased/pareto-latest","description":"This model always redirects to the latest model in the Pareto family.","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"~unbiased","completion_ratio":4,"cache_ratio":0.0375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~unbiased","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"~unbiased/pareto-latest","context_length":1048576}},{"model_name":"aion-2.0","description":"Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling. It is particularly strong at introducing tension, crises, and conflict into stories, making narratives feel more engaging....","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1.5999999999999999,"openrouter_id":"aion-labs/aion-2.0","context_length":131072}},{"model_name":"aion-labs/aion-2.0","description":"Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling. It is particularly strong at introducing tension, crises, and conflict into stories, making narratives feel more engaging....","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1.5999999999999999,"openrouter_id":"aion-labs/aion-2.0","context_length":131072}},{"model_name":"aion-labs/aion-rp-llama-3.1-8b","description":"Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses. It is a fine-tuned base model...","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1.5999999999999999,"openrouter_id":"aion-labs/aion-rp-llama-3.1-8b","context_length":32768}},{"model_name":"aion-rp-llama-3.1-8b","description":"Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses. It is a fine-tuned base model...","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1.5999999999999999,"openrouter_id":"aion-labs/aion-rp-llama-3.1-8b","context_length":32768}},{"model_name":"amazon/nova-pro-v1","description":"Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks. As of December...","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"amazon","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"amazon/nova-pro-v1","context_length":300000}},{"model_name":"morph-v3-fast","description":"Morph's fastest apply model for code edits. ~10,500 tokens/sec with 96% accuracy for rapid code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code> <update>{edit_snippet}</update>...","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"morph","completion_ratio":1.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,morph","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1.2,"openrouter_id":"morph/morph-v3-fast","context_length":262144}},{"model_name":"morph/morph-v3-fast","description":"Morph's fastest apply model for code edits. ~10,500 tokens/sec with 96% accuracy for rapid code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code> <update>{edit_snippet}</update>...","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"morph","completion_ratio":1.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,morph","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1.2,"openrouter_id":"morph/morph-v3-fast","context_length":262144}},{"model_name":"nova-pro-v1","description":"Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks. As of December...","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"amazon","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"amazon/nova-pro-v1","context_length":300000}},{"model_name":"pareto-26.10","description":"Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"unbiased","completion_ratio":4,"cache_ratio":0.0375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,unbiased","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"unbiased/pareto-26.10","context_length":1048576}},{"model_name":"pareto-latest","description":"This model always redirects to the latest model in the Pareto family.","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"~unbiased","completion_ratio":4,"cache_ratio":0.0375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~unbiased","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"~unbiased/pareto-latest","context_length":1048576}},{"model_name":"qwen/qwen2.5-vl-72b-instruct","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.","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"qwen","completion_ratio":1.25,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1,"openrouter_id":"qwen/qwen2.5-vl-72b-instruct","context_length":128000}},{"model_name":"qwen2.5-vl-72b-instruct","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.","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"qwen","completion_ratio":1.25,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":1,"openrouter_id":"qwen/qwen2.5-vl-72b-instruct","context_length":128000}},{"model_name":"unbiased/pareto-26.10","description":"Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"unbiased","completion_ratio":4,"cache_ratio":0.0375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,unbiased","grokcode":{"input_usd_per_m":0.7999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"unbiased/pareto-26.10","context_length":1048576}},{"model_name":"claude-3-5-haiku-20241022","description":"claude-3-5-haiku-20241022","quota_type":0,"model_ratio":0.4,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":0.8,"output_usd_per_m":4,"context_length":null}},{"model_name":"gemini-3.8-flash","description":"Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.","quota_type":0,"model_ratio":0.4125,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.050505,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.8250000000000001,"output_usd_per_m":4.125,"openrouter_id":"google/gemini-3.8-flash","context_length":1048576}},{"model_name":"google/gemini-3.8-flash","description":"Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.","quota_type":0,"model_ratio":0.4125,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":0.050505,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0.8250000000000001,"output_usd_per_m":4.125,"openrouter_id":"google/gemini-3.8-flash","context_length":1048576}},{"model_name":"l3.1-euryale-70b","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).","quota_type":0,"model_ratio":0.425,"model_price":0,"owner_by":"sao10k","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sao10k","grokcode":{"input_usd_per_m":0.85,"output_usd_per_m":0.85,"openrouter_id":"sao10k/l3.1-euryale-70b","context_length":131072}},{"model_name":"relace-apply-3","description":"Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files. It can apply updates from GPT-4o, Claude, and others into your files at...","quota_type":0,"model_ratio":0.425,"model_price":0,"owner_by":"relace","completion_ratio":1.4706,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,relace","grokcode":{"input_usd_per_m":0.85,"output_usd_per_m":1.25,"openrouter_id":"relace/relace-apply-3","context_length":256000}},{"model_name":"relace/relace-apply-3","description":"Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files. It can apply updates from GPT-4o, Claude, and others into your files at...","quota_type":0,"model_ratio":0.425,"model_price":0,"owner_by":"relace","completion_ratio":1.4706,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,relace","grokcode":{"input_usd_per_m":0.85,"output_usd_per_m":1.25,"openrouter_id":"relace/relace-apply-3","context_length":256000}},{"model_name":"sao10k/l3.1-euryale-70b","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).","quota_type":0,"model_ratio":0.425,"model_price":0,"owner_by":"sao10k","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sao10k","grokcode":{"input_usd_per_m":0.85,"output_usd_per_m":0.85,"openrouter_id":"sao10k/l3.1-euryale-70b","context_length":131072}},{"model_name":"openai/gpt-5.2:batch","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...","quota_type":0,"model_ratio":0.4375,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":0.875,"output_usd_per_m":7,"openrouter_id":"openai/gpt-5.2:batch","context_length":400000}},{"model_name":"morph-v3-large","description":"Morph's high-accuracy apply model for complex code edits. ~4,500 tokens/sec with 98% accuracy for precise code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code>...","quota_type":0,"model_ratio":0.45,"model_price":0,"owner_by":"morph","completion_ratio":2.1111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,morph","grokcode":{"input_usd_per_m":0.8999999999999999,"output_usd_per_m":1.9,"openrouter_id":"morph/morph-v3-large","context_length":262144}},{"model_name":"morph/morph-v3-large","description":"Morph's high-accuracy apply model for complex code edits. ~4,500 tokens/sec with 98% accuracy for precise code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code>...","quota_type":0,"model_ratio":0.45,"model_price":0,"owner_by":"morph","completion_ratio":2.1111,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,morph","grokcode":{"input_usd_per_m":0.8999999999999999,"output_usd_per_m":1.9,"openrouter_id":"morph/morph-v3-large","context_length":262144}},{"model_name":"Kimi K2.6","description":"Kimi 新一代长程代码和 Agent 模型，支持文本、图片与视频输入。","quota_type":0,"model_ratio":0.475,"model_price":0,"owner_by":"Kimi","completion_ratio":4.2105,"cache_ratio":0.168421,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Kimi","grokcode":{"input_usd_per_m":0.95,"output_usd_per_m":4,"context_length":null}},{"model_name":"sakana-namazu","description":"Sakana Namazu is a Japanese-specialized reasoning model from Sakana AI, based on Kimi K2.6 with additional training for Japanese language and business contexts. It is suited for Japanese instruction following,...","quota_type":0,"model_ratio":0.475,"model_price":0,"owner_by":"sakana","completion_ratio":4.2105,"cache_ratio":0.157895,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":0.95,"output_usd_per_m":4,"openrouter_id":"sakana/sakana-namazu","context_length":262144}},{"model_name":"sakana/sakana-namazu","description":"Sakana Namazu is a Japanese-specialized reasoning model from Sakana AI, based on Kimi K2.6 with additional training for Japanese language and business contexts. It is suited for Japanese instruction following,...","quota_type":0,"model_ratio":0.475,"model_price":0,"owner_by":"sakana","completion_ratio":4.2105,"cache_ratio":0.157895,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":0.95,"output_usd_per_m":4,"openrouter_id":"sakana/sakana-namazu","context_length":262144}},{"model_name":"~openai/gpt-terra-latest","description":"This model always redirects to the latest model in the GPT Terra family.","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"~openai/gpt-terra-latest","context_length":1050000}},{"model_name":"anthropic/claude-haiku-4.5","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...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"anthropic/claude-haiku-4.5","context_length":200000}},{"model_name":"anthropic/claude-sonnet-5:batch","description":"Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"anthropic/claude-sonnet-5:batch","context_length":1000000}},{"model_name":"anthropic/claude-sonnet-5.5:batch","description":"Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"anthropic/claude-sonnet-5.5:batch","context_length":1000000}},{"model_name":"Claude Haiku 4.5","description":"Anthropic Claude 低延迟低成本模型，适合轻量改写、抽取、分类和工具调用前置判断。","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"Claude","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,Claude","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"context_length":null}},{"model_name":"claude-haiku-4-5-20251001","description":"claude-haiku-4-5-20251001","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"context_length":null}},{"model_name":"claude-haiku-4.5","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...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"anthropic/claude-haiku-4.5","context_length":200000}},{"model_name":"glm-5","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...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"glm","completion_ratio":3.2,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1,"output_usd_per_m":3.1999999999999997,"openrouter_id":"z-ai/glm-5","context_length":204800}},{"model_name":"GLM-5","description":"智谱 GLM 基座模型，面向复杂系统工程与长程 Agent 任务。","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"GLM","completion_ratio":3.2,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,GLM","grokcode":{"input_usd_per_m":1,"output_usd_per_m":3.2,"context_length":null}},{"model_name":"google/gemini-3.1-pro-preview:batch","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...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1,"output_usd_per_m":6,"openrouter_id":"google/gemini-3.1-pro-preview:batch","context_length":1048576}},{"model_name":"google/gemini-4-argon:batch","description":"Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"google/gemini-4-argon:batch","context_length":1048576}},{"model_name":"gpt-3.5-turbo-0613","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.","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2,"openrouter_id":"openai/gpt-3.5-turbo-0613","context_length":4095}},{"model_name":"GPT-5.6 Luna","description":"GPT-5.6 Luna","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":6,"context_length":null}},{"model_name":"GPT-6 Luna","description":"GPT-6 Luna","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":6,"context_length":null}},{"model_name":"gpt-6.1-terra","description":"GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6.1-terra","context_length":1050000}},{"model_name":"gpt-6.1-terra-pro","description":"GPT-6.1 Terra Pro is the same underlying model as [GPT-6.1 Terra](https://openrouter.ai/openai/gpt-6.1-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6.1-terra-pro","context_length":1050000}},{"model_name":"gpt-terra-latest","description":"This model always redirects to the latest model in the GPT Terra family.","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"~openai/gpt-terra-latest","context_length":1050000}},{"model_name":"Grok Build 0.1","description":"Grok Build 0.1","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"xai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2,"context_length":null}},{"model_name":"grok-build-0.1","description":"Grok Build 0.1 is SpaceXAI’s fast coding model trained specifically for agentic software engineering workflows. It supports text and image inputs with text output, and is optimized for interactive coding...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2,"openrouter_id":"x-ai/grok-build-0.1","context_length":256000}},{"model_name":"hermes-4-405b","description":"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"nousresearch","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nousresearch","grokcode":{"input_usd_per_m":1,"output_usd_per_m":3,"openrouter_id":"nousresearch/hermes-4-405b","context_length":131072}},{"model_name":"inkling","description":"Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems,...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"thinkingmachines","completion_ratio":4.05,"cache_ratio":0.17,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thinkingmachines","grokcode":{"input_usd_per_m":1,"output_usd_per_m":4.05,"openrouter_id":"thinkingmachines/inkling","context_length":524288}},{"model_name":"nousresearch/hermes-4-405b","description":"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"nousresearch","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nousresearch","grokcode":{"input_usd_per_m":1,"output_usd_per_m":3,"openrouter_id":"nousresearch/hermes-4-405b","context_length":131072}},{"model_name":"openai/gpt-3.5-turbo-0613","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.","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2,"openrouter_id":"openai/gpt-3.5-turbo-0613","context_length":4095}},{"model_name":"openai/gpt-4.1:batch","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...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":4,"openrouter_id":"openai/gpt-4.1:batch","context_length":1047576}},{"model_name":"openai/gpt-5.6-terra-pro:batch","description":"GPT-5.6 Terra Pro is the same underlying model as [GPT-5.6 Terra](https://openrouter.ai/openai/gpt-5.6-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":6,"openrouter_id":"openai/gpt-5.6-terra-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-5.6-terra:batch","description":"GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier. It is suited for everyday coding, reasoning, and agentic...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":6,"openrouter_id":"openai/gpt-5.6-terra:batch","context_length":1050000}},{"model_name":"openai/gpt-6-sol-pro:batch","description":"GPT-6 Sol Pro is the same underlying model as [GPT-6 Sol](https://openrouter.ai/openai/gpt-6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6-sol-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-6-sol:batch","description":"GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6-sol:batch","context_length":1050000}},{"model_name":"openai/gpt-6.1-sol-pro:batch","description":"GPT-6.1 Sol Pro is the same underlying model as [GPT-6.1 Sol](https://openrouter.ai/openai/gpt-6.1-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6.1-sol-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-6.1-sol:batch","description":"GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6.1-sol:batch","context_length":1050000}},{"model_name":"openai/gpt-6.1-terra","description":"GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6.1-terra","context_length":1050000}},{"model_name":"openai/gpt-6.1-terra-pro","description":"GPT-6.1 Terra Pro is the same underlying model as [GPT-6.1 Terra](https://openrouter.ai/openai/gpt-6.1-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":5,"openrouter_id":"openai/gpt-6.1-terra-pro","context_length":1050000}},{"model_name":"openai/o3:batch","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....","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":4,"openrouter_id":"openai/o3:batch","context_length":200000}},{"model_name":"perplexity/sonar","description":"Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources. It is designed for companies seeking to integrate lightweight question-and-answer features...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"perplexity","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":1,"output_usd_per_m":1,"openrouter_id":"perplexity/sonar","context_length":127072}},{"model_name":"relace-search","description":"The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request. In contrast to RAG, relace-search performs agentic...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"relace","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,relace","grokcode":{"input_usd_per_m":1,"output_usd_per_m":3,"openrouter_id":"relace/relace-search","context_length":256000}},{"model_name":"relace/relace-search","description":"The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request. In contrast to RAG, relace-search performs agentic...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"relace","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,relace","grokcode":{"input_usd_per_m":1,"output_usd_per_m":3,"openrouter_id":"relace/relace-search","context_length":256000}},{"model_name":"sonar","description":"Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources. It is designed for companies seeking to integrate lightweight question-and-answer features...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"perplexity","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":1,"output_usd_per_m":1,"openrouter_id":"perplexity/sonar","context_length":127072}},{"model_name":"step-5-preview","description":"Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"stepfun","completion_ratio":2.7,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stepfun","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2.7,"openrouter_id":"stepfun/step-5-preview","context_length":1048576}},{"model_name":"stepfun/step-5-preview","description":"Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"stepfun","completion_ratio":2.7,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stepfun","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2.7,"openrouter_id":"stepfun/step-5-preview","context_length":1048576}},{"model_name":"thinkingmachines/inkling","description":"Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems,...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"thinkingmachines","completion_ratio":4.05,"cache_ratio":0.17,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,thinkingmachines","grokcode":{"input_usd_per_m":1,"output_usd_per_m":4.05,"openrouter_id":"thinkingmachines/inkling","context_length":524288}},{"model_name":"x-ai/grok-4.3:batch","description":"Grok 4.3 is a reasoning model from SpaceXAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2,"openrouter_id":"x-ai/grok-4.3:batch","context_length":1000000}},{"model_name":"x-ai/grok-build-0.1","description":"Grok Build 0.1 is SpaceXAI’s fast coding model trained specifically for agentic software engineering workflows. It supports text and image inputs with text output, and is optimized for interactive coding...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1,"output_usd_per_m":2,"openrouter_id":"x-ai/grok-build-0.1","context_length":256000}},{"model_name":"z-ai/glm-5","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...","quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"glm","completion_ratio":3.2,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1,"output_usd_per_m":3.1999999999999997,"openrouter_id":"z-ai/glm-5","context_length":204800}},{"model_name":"gpt-4o-realtime-preview","description":"gpt-4o-realtime-preview","quota_type":0,"model_ratio":0.555556,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":1.11111111,"output_usd_per_m":4.44444444,"context_length":null}},{"model_name":"glm-5-turbo","description":"GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios. It is deeply optimized for real-world agent workflows...","quota_type":0,"model_ratio":0.6,"model_price":0,"owner_by":"glm","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1.2,"output_usd_per_m":4,"openrouter_id":"z-ai/glm-5-turbo","context_length":202752}},{"model_name":"glm-5v-turbo","description":"GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks. It natively handles image, video, and text inputs, excels at long-horizon planning, complex coding,...","quota_type":0,"model_ratio":0.6,"model_price":0,"owner_by":"glm","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1.2,"output_usd_per_m":4,"openrouter_id":"z-ai/glm-5v-turbo","context_length":202752}},{"model_name":"z-ai/glm-5-turbo","description":"GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios. It is deeply optimized for real-world agent workflows...","quota_type":0,"model_ratio":0.6,"model_price":0,"owner_by":"glm","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1.2,"output_usd_per_m":4,"openrouter_id":"z-ai/glm-5-turbo","context_length":202752}},{"model_name":"z-ai/glm-5v-turbo","description":"GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks. It natively handles image, video, and text inputs, excels at long-horizon planning, complex coding,...","quota_type":0,"model_ratio":0.6,"model_price":0,"owner_by":"glm","completion_ratio":3.3333,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1.2,"output_usd_per_m":4,"openrouter_id":"z-ai/glm-5v-turbo","context_length":202752}},{"model_name":"gemini-2.5-pro","description":"长上下文与推理","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"gemini","completion_ratio":8,"cache_ratio":0.252,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,gemini","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"context_length":null}},{"model_name":"gemini-2.5-pro-preview","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"google","completion_ratio":8,"cache_ratio":0.1,"create_cache_ratio":0.3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"google/gemini-2.5-pro-preview","context_length":1048576}},{"model_name":"google/gemini-2.5-pro","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"google","completion_ratio":8,"cache_ratio":0.1,"create_cache_ratio":0.3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"google/gemini-2.5-pro","context_length":1048576}},{"model_name":"google/gemini-2.5-pro-preview","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"google","completion_ratio":8,"cache_ratio":0.1,"create_cache_ratio":0.3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"google/gemini-2.5-pro-preview","context_length":1048576}},{"model_name":"gpt-5","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5","context_length":400000}},{"model_name":"gpt-5.1","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.1","context_length":400000}},{"model_name":"gpt-5.1-codex","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....","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.104,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.1-codex","context_length":400000}},{"model_name":"gpt-5.1-codex-max","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.1-codex-max","context_length":400000}},{"model_name":"Grok 4.20","description":"xAI Grok 文本模型，适合推理、代码和 Agent 任务。","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"Grok","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Grok","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"context_length":null}},{"model_name":"Grok 4.20 Multi Agent","description":"Grok 4.20 Multi Agent","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"context_length":null}},{"model_name":"Grok 4.20 Non Reasoning","description":"Grok 4.20 Non Reasoning","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"context_length":null}},{"model_name":"Grok 4.20 Reasoning","description":"Grok 4.20 Reasoning","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"context_length":null}},{"model_name":"Grok 4.3","description":"Grok 4.3","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"context_length":null}},{"model_name":"grok-4.20","description":"Grok 4.20 is a reasoning model from SpaceXAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherance, delivering...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"openrouter_id":"x-ai/grok-4.20","context_length":2000000}},{"model_name":"grok-4.20-multi-agent","description":"Grok 4.20 Multi-Agent is a variant of SpaceXAI’s Grok 4.20 designed for collaborative, agent-based workflows. Multiple agents operate in parallel to conduct deep research, coordinate tool use, and synthesize information...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"openrouter_id":"x-ai/grok-4.20-multi-agent","context_length":2000000}},{"model_name":"grok-4.3","description":"Grok 4.3 is a reasoning model from SpaceXAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"openrouter_id":"x-ai/grok-4.3","context_length":1000000}},{"model_name":"meta/muse-spark-1.1","description":"Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. It accepts text, images, video, and PDF documents and returns text, with a 1M-token context window....","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-1.1","context_length":1048576}},{"model_name":"meta/muse-spark-1.2","description":"Muse Spark 1.2 is a reasoning model from Meta, designed for complex agentic tasks. It accepts text, images, video, and PDF documents, returns text, and offers a 1M-token context window....","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-1.2","context_length":1048576}},{"model_name":"meta/muse-spark-1.3","description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It is designed to keep track of information across extended tasks, work through...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-1.3","context_length":1048576}},{"model_name":"meta/muse-spark-open","description":"Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-open","context_length":1048576}},{"model_name":"muse-spark-1.1","description":"Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. It accepts text, images, video, and PDF documents and returns text, with a 1M-token context window....","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-1.1","context_length":1048576}},{"model_name":"muse-spark-1.2","description":"Muse Spark 1.2 is a reasoning model from Meta, designed for complex agentic tasks. It accepts text, images, video, and PDF documents, returns text, and offers a 1M-token context window....","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-1.2","context_length":1048576}},{"model_name":"muse-spark-1.3","description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It is designed to keep track of information across extended tasks, work through...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-1.3","context_length":1048576}},{"model_name":"muse-spark-open","description":"Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"llama","completion_ratio":3.4,"cache_ratio":0.12,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":4.25,"openrouter_id":"meta/muse-spark-open","context_length":1048576}},{"model_name":"openai/gpt-4o:batch","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":5,"openrouter_id":"openai/gpt-4o:batch","context_length":128000}},{"model_name":"openai/gpt-5","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5","context_length":400000}},{"model_name":"openai/gpt-5.1","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.1","context_length":400000}},{"model_name":"openai/gpt-5.1-codex","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....","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.104,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.1-codex","context_length":400000}},{"model_name":"openai/gpt-5.1-codex-max","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.1-codex-max","context_length":400000}},{"model_name":"openai/gpt-5.4:batch","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...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":7.5,"openrouter_id":"openai/gpt-5.4:batch","context_length":1050000}},{"model_name":"x-ai/grok-4.20","description":"Grok 4.20 is a reasoning model from SpaceXAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherance, delivering...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"openrouter_id":"x-ai/grok-4.20","context_length":2000000}},{"model_name":"x-ai/grok-4.20-multi-agent","description":"Grok 4.20 Multi-Agent is a variant of SpaceXAI’s Grok 4.20 designed for collaborative, agent-based workflows. Multiple agents operate in parallel to conduct deep research, coordinate tool use, and synthesize information...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"openrouter_id":"x-ai/grok-4.20-multi-agent","context_length":2000000}},{"model_name":"x-ai/grok-4.3","description":"Grok 4.3 is a reasoning model from SpaceXAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual...","quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"xai","completion_ratio":2,"cache_ratio":0.16,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.25,"output_usd_per_m":2.5,"openrouter_id":"x-ai/grok-4.3","context_length":1000000}},{"model_name":"glm-5.1","description":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks. Unlike previous models built around minute-level interactions, GLM-5.1 can work independently and continuously on...","quota_type":0,"model_ratio":0.7,"model_price":0,"owner_by":"glm","completion_ratio":3.1429,"cache_ratio":0.185714,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1.4,"output_usd_per_m":4.4,"openrouter_id":"z-ai/glm-5.1","context_length":204800}},{"model_name":"GLM-5.1","description":"GLM 5 系列高能力版本，已补齐智谱开放平台官方按量价格，适合长程 Agent 和工程级任务。","quota_type":0,"model_ratio":0.7,"model_price":0,"owner_by":"GLM","completion_ratio":3.1429,"cache_ratio":0.185714,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,GLM","grokcode":{"input_usd_per_m":1.4,"output_usd_per_m":4.4,"context_length":null}},{"model_name":"z-ai/glm-5.1","description":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks. Unlike previous models built around minute-level interactions, GLM-5.1 can work independently and continuously on...","quota_type":0,"model_ratio":0.7,"model_price":0,"owner_by":"glm","completion_ratio":3.1429,"cache_ratio":0.185714,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":1.4,"output_usd_per_m":4.4,"openrouter_id":"z-ai/glm-5.1","context_length":204800}},{"model_name":"qwen/qwen3.7-max","description":"Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks,...","quota_type":0,"model_ratio":0.7375,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":1.475,"output_usd_per_m":4.425,"openrouter_id":"qwen/qwen3.7-max","context_length":1000000}},{"model_name":"qwen3.7-max","description":"Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks,...","quota_type":0,"model_ratio":0.7375,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":1.475,"output_usd_per_m":4.425,"openrouter_id":"qwen/qwen3.7-max","context_length":1000000}},{"model_name":"anthropic/claude-sonnet-4.5:batch","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...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":7.5,"openrouter_id":"anthropic/claude-sonnet-4.5:batch","context_length":1000000}},{"model_name":"anthropic/claude-sonnet-4.6:batch","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...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":7.5,"openrouter_id":"anthropic/claude-sonnet-4.6:batch","context_length":1000000}},{"model_name":"Gemini 3.5 Flash","description":"Google Gemini Flash 模型，适合低延迟、多模态和高吞吐任务。","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"Gemini","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Gemini","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":9,"context_length":null}},{"model_name":"gemini-3.5-flash","description":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":9,"openrouter_id":"google/gemini-3.5-flash","context_length":1048576}},{"model_name":"gemini-nano-banana-2.1","description":"Nano Banana 2.1 (Gemini Nano Banana 2.1) is Google's image generation and editing model on the Flash tier, succeeding Nano Banana 2 and Nano Banana Pro. It improves product recontextualization,...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":7.5,"openrouter_id":"google/gemini-nano-banana-2.1","context_length":65536}},{"model_name":"gemini-omni-1.1-flash","description":"Gemini Omni 1.1 Flash is Google's native multimodal video generation and editing model, the generally available successor to Gemini Omni Flash Preview. It is suited for text-to-video, image-to-video, first and...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":9,"openrouter_id":"google/gemini-omni-1.1-flash","context_length":131072}},{"model_name":"gemini-omni-1.2-flash","description":"Gemini Omni 1.2 Flash is Google's native multimodal video generation and editing model, the successor to Gemini Omni 1.1 Flash. It generates clips up to 15 seconds long with native...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":9,"openrouter_id":"google/gemini-omni-1.2-flash","context_length":131072}},{"model_name":"google/gemini-3.5-flash","description":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.055556,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":9,"openrouter_id":"google/gemini-3.5-flash","context_length":1048576}},{"model_name":"google/gemini-nano-banana-2.1","description":"Nano Banana 2.1 (Gemini Nano Banana 2.1) is Google's image generation and editing model on the Flash tier, succeeding Nano Banana 2 and Nano Banana Pro. It improves product recontextualization,...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":7.5,"openrouter_id":"google/gemini-nano-banana-2.1","context_length":65536}},{"model_name":"google/gemini-omni-1.1-flash","description":"Gemini Omni 1.1 Flash is Google's native multimodal video generation and editing model, the generally available successor to Gemini Omni Flash Preview. It is suited for text-to-video, image-to-video, first and...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":9,"openrouter_id":"google/gemini-omni-1.1-flash","context_length":131072}},{"model_name":"google/gemini-omni-1.2-flash","description":"Gemini Omni 1.2 Flash is Google's native multimodal video generation and editing model, the successor to Gemini Omni 1.1 Flash. It generates clips up to 15 seconds long with native...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"google","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":9,"openrouter_id":"google/gemini-omni-1.2-flash","context_length":131072}},{"model_name":"mistral-medium-3-5","description":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"mistral","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":7.5,"openrouter_id":"mistralai/mistral-medium-3-5","context_length":262144}},{"model_name":"mistralai/mistral-medium-3-5","description":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...","quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"mistral","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":1.5,"output_usd_per_m":7.5,"openrouter_id":"mistralai/mistral-medium-3-5","context_length":262144}},{"model_name":"deepseek-v4-pro","description":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...","quota_type":0,"model_ratio":0.8,"model_price":0,"owner_by":"deepseek","completion_ratio":2,"cache_ratio":0.084375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":1.5999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"deepseek/deepseek-v4-pro","context_length":1048576}},{"model_name":"deepseek/deepseek-v4-pro","description":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...","quota_type":0,"model_ratio":0.8,"model_price":0,"owner_by":"deepseek","completion_ratio":2,"cache_ratio":0.084375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,deepseek","grokcode":{"input_usd_per_m":1.5999999999999999,"output_usd_per_m":3.1999999999999997,"openrouter_id":"deepseek/deepseek-v4-pro","context_length":1048576}},{"model_name":"x-ai/grok-4.6:batch","description":"Grok 4.6 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7).","quota_type":0,"model_ratio":0.8,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.5999999999999999,"output_usd_per_m":4.8,"openrouter_id":"x-ai/grok-4.6:batch","context_length":500000}},{"model_name":"x-ai/grok-4.7:batch","description":"Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...","quota_type":0,"model_ratio":0.8,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":1.5999999999999999,"output_usd_per_m":4.8,"openrouter_id":"x-ai/grok-4.7:batch","context_length":500000}},{"model_name":"qwen-max","description":"通义旗舰","quota_type":0,"model_ratio":0.8,"model_price":0,"owner_by":"qwen","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,qwen","grokcode":{"input_usd_per_m":1.6,"output_usd_per_m":6.4,"context_length":null}},{"model_name":"gpt-5.2","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...","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"openrouter_id":"openai/gpt-5.2","context_length":400000}},{"model_name":"GPT-5.2","description":"GPT-5.2","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"context_length":null}},{"model_name":"gpt-5.2-2025-12-11","description":"gpt-5.2-2025-12-11","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"context_length":null}},{"model_name":"gpt-5.2-chat","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...","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"openrouter_id":"openai/gpt-5.2-chat","context_length":128000}},{"model_name":"gpt-5.2-chat-latest","description":"gpt-5.2-chat-latest","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"context_length":null}},{"model_name":"gpt-5.2-codex","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....","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"openrouter_id":"openai/gpt-5.2-codex","context_length":400000}},{"model_name":"GPT-5.3 Codex Spark","description":"GPT-5.3 Codex Spark","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"context_length":null}},{"model_name":"gpt-5.3-codex","description":"gpt-5.3-codex","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"context_length":null}},{"model_name":"openai/gpt-5.2","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...","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"openrouter_id":"openai/gpt-5.2","context_length":400000}},{"model_name":"openai/gpt-5.2-chat","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...","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"openrouter_id":"openai/gpt-5.2-chat","context_length":128000}},{"model_name":"openai/gpt-5.2-codex","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....","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"openrouter_id":"openai/gpt-5.2-codex","context_length":400000}},{"model_name":"openai/gpt-5.3-codex","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...","quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"openai","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":1.75,"output_usd_per_m":14,"openrouter_id":"openai/gpt-5.3-codex","context_length":400000}},{"model_name":"~anthropic/claude-sonnet-latest","description":"This model always redirects to the latest model in the Claude Sonnet family.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"~anthropic/claude-sonnet-latest","context_length":1000000}},{"model_name":"~google/gemini-pro-latest","description":"This model always redirects to the latest model in the Gemini Pro family.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~google","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":0.09375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"~google/gemini-pro-latest","context_length":1048576}},{"model_name":"~openai/gpt-sol-latest","description":"This model always redirects to the latest model in the GPT Sol family.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"~openai/gpt-sol-latest","context_length":1050000}},{"model_name":"~x-ai/grok-latest","description":"This model always redirects to the latest Grok model from xAI.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~x-ai","completion_ratio":3,"cache_ratio":0.25,"create_cache_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~x-ai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"~x-ai/grok-latest","context_length":500000}},{"model_name":"anthropic/claude-opus-5.5:batch","description":"Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"anthropic/claude-opus-5.5:batch","context_length":1000000}},{"model_name":"anthropic/claude-sonnet-5","description":"Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"anthropic/claude-sonnet-5","context_length":1000000}},{"model_name":"anthropic/claude-sonnet-5.5","description":"Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"anthropic/claude-sonnet-5.5","context_length":1000000}},{"model_name":"Claude Sonnet 5","description":"Anthropic Claude 均衡主力模型，适合日常开发、代码修改、长上下文阅读和 Agent 工作流。","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"Claude","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,Claude","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"context_length":null}},{"model_name":"claude-sonnet-5","description":"Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"anthropic/claude-sonnet-5","context_length":1000000}},{"model_name":"claude-sonnet-5.5","description":"Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"anthropic/claude-sonnet-5.5","context_length":1000000}},{"model_name":"claude-sonnet-latest","description":"This model always redirects to the latest model in the Claude Sonnet family.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"~anthropic/claude-sonnet-latest","context_length":1000000}},{"model_name":"fugu-max","description":"Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"sakana","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"sakana/fugu-max","context_length":1000000}},{"model_name":"Gemini 3.1 Pro Preview","description":"Google Gemini 高能力预览模型，适合多模态理解、长上下文和复杂推理任务。","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"Gemini","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Gemini","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"context_length":null}},{"model_name":"gemini-3-pro-image","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3-pro-image","context_length":131072}},{"model_name":"gemini-3-pro-image-preview","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3-pro-image-preview","context_length":65536}},{"model_name":"gemini-3.1-pro-preview","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3.1-pro-preview","context_length":1048576}},{"model_name":"gemini-3.1-pro-preview-customtools","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3.1-pro-preview-customtools","context_length":1048576}},{"model_name":"gemini-4-argon","description":"Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":0.09375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"google/gemini-4-argon","context_length":1048576}},{"model_name":"gemini-pro-latest","description":"This model always redirects to the latest model in the Gemini Pro family.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~google","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":0.09375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"~google/gemini-pro-latest","context_length":1048576}},{"model_name":"google/gemini-3-pro-image","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3-pro-image","context_length":131072}},{"model_name":"google/gemini-3-pro-image-preview","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3-pro-image-preview","context_length":65536}},{"model_name":"google/gemini-3.1-pro-preview","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3.1-pro-preview","context_length":1048576}},{"model_name":"google/gemini-3.1-pro-preview-customtools","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":0.1875,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"google/gemini-3.1-pro-preview-customtools","context_length":1048576}},{"model_name":"google/gemini-4-argon","description":"Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"google","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":0.09375,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"google/gemini-4-argon","context_length":1048576}},{"model_name":"gpt-4.1","description":"gpt-4.1","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"context_length":null}},{"model_name":"GPT-5.6 Terra","description":"OpenAI 高性价比通用模型，适合日常 Coding Agent 与产品内自动化。","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"OpenAI","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,OpenAI","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"context_length":null}},{"model_name":"gpt-5.6-terra","description":"GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier. It is suited for everyday coding, reasoning, and agentic...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"openai/gpt-5.6-terra","context_length":1050000}},{"model_name":"gpt-5.6-terra-openai-compact","description":"gpt-5.6-terra-openai-compact","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"context_length":null}},{"model_name":"gpt-5.6-terra-pro","description":"GPT-5.6 Terra Pro is the same underlying model as [GPT-5.6 Terra](https://openrouter.ai/openai/gpt-5.6-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"openai/gpt-5.6-terra-pro","context_length":1050000}},{"model_name":"GPT-6 Sol","description":"GPT-6 Sol","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"context_length":null}},{"model_name":"gpt-6-sol","description":"GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6-sol","context_length":1050000}},{"model_name":"gpt-6-sol-pro","description":"GPT-6 Sol Pro is the same underlying model as [GPT-6 Sol](https://openrouter.ai/openai/gpt-6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6-sol-pro","context_length":1050000}},{"model_name":"GPT-6.1 Sol","description":"GPT-6.1 Sol","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"context_length":null}},{"model_name":"GPT-6.1 Terra","description":"GPT-6.1 Terra","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":3.75,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":7.5,"context_length":null}},{"model_name":"gpt-6.1-sol","description":"GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6.1-sol","context_length":1050000}},{"model_name":"gpt-6.1-sol-pro","description":"GPT-6.1 Sol Pro is the same underlying model as [GPT-6.1 Sol](https://openrouter.ai/openai/gpt-6.1-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6.1-sol-pro","context_length":1050000}},{"model_name":"gpt-sol-latest","description":"This model always redirects to the latest model in the GPT Sol family.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"~openai/gpt-sol-latest","context_length":1050000}},{"model_name":"Grok 4.5","description":"Grok 4.5","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"context_length":null}},{"model_name":"Grok 4.6","description":"Grok 4.6","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"context_length":null}},{"model_name":"Grok 4.7","description":"Grok 4.7","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"context_length":null}},{"model_name":"grok-4.5","description":"Grok 4.5 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.15,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"x-ai/grok-4.5","context_length":500000}},{"model_name":"grok-4.6","description":"Grok 4.6 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7).","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"x-ai/grok-4.6","context_length":500000}},{"model_name":"grok-4.7","description":"Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"x-ai/grok-4.7","context_length":500000}},{"model_name":"grok-latest","description":"This model always redirects to the latest Grok model from xAI.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"~x-ai","completion_ratio":3,"cache_ratio":0.25,"create_cache_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~x-ai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"~x-ai/grok-latest","context_length":500000}},{"model_name":"kimi-k3.1-pro","description":"Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"kimi","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"moonshotai/kimi-k3.1-pro","context_length":1048576}},{"model_name":"mistral-large","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/)....","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"mistralai/mistral-large","context_length":128000}},{"model_name":"mistral-large-2407","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/)....","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"mistralai/mistral-large-2407","context_length":131072}},{"model_name":"mistralai/mistral-large","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/)....","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"mistralai/mistral-large","context_length":128000}},{"model_name":"mistralai/mistral-large-2407","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/)....","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"mistralai/mistral-large-2407","context_length":131072}},{"model_name":"mistralai/mixtral-8x22b-instruct","description":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include: - strong math, coding,...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"mistralai/mixtral-8x22b-instruct","context_length":65536}},{"model_name":"mixtral-8x22b-instruct","description":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include: - strong math, coding,...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"mistral","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,mistral","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"mistralai/mixtral-8x22b-instruct","context_length":65536}},{"model_name":"moonshotai/kimi-k3.1-pro","description":"Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"kimi","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"moonshotai/kimi-k3.1-pro","context_length":1048576}},{"model_name":"Nano Banana Pro","description":"Google Gemini 3 Pro Image 图片生成模型，适合高质量图片生成、编辑和多模态图像工作流。","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"图片生成","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,图片生成","grokcode":{"input_usd_per_m":2,"output_usd_per_m":0,"context_length":null}},{"model_name":"o3","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....","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"openrouter_id":"openai/o3","context_length":200000}},{"model_name":"openai/gpt-4.1","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...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"openrouter_id":"openai/gpt-4.1","context_length":1047576}},{"model_name":"openai/gpt-5.6-sol-pro:batch","description":"GPT-5.6 Sol Pro is the same underlying model as [GPT-5.6 Sol](https://openrouter.ai/openai/gpt-5.6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.6-sol-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-5.6-sol:batch","description":"GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5.6-sol:batch","context_length":1050000}},{"model_name":"openai/gpt-5.6-terra","description":"GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier. It is suited for everyday coding, reasoning, and agentic...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"openai/gpt-5.6-terra","context_length":1050000}},{"model_name":"openai/gpt-5.6-terra-pro","description":"GPT-5.6 Terra Pro is the same underlying model as [GPT-5.6 Terra](https://openrouter.ai/openai/gpt-5.6-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":12,"openrouter_id":"openai/gpt-5.6-terra-pro","context_length":1050000}},{"model_name":"openai/gpt-6-sol","description":"GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6-sol","context_length":1050000}},{"model_name":"openai/gpt-6-sol-pro","description":"GPT-6 Sol Pro is the same underlying model as [GPT-6 Sol](https://openrouter.ai/openai/gpt-6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6-sol-pro","context_length":1050000}},{"model_name":"openai/gpt-6.1-sol","description":"GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6.1-sol","context_length":1050000}},{"model_name":"openai/gpt-6.1-sol-pro","description":"GPT-6.1 Sol Pro is the same underlying model as [GPT-6.1 Sol](https://openrouter.ai/openai/gpt-6.1-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":10,"openrouter_id":"openai/gpt-6.1-sol-pro","context_length":1050000}},{"model_name":"openai/o3","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....","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"openrouter_id":"openai/o3","context_length":200000}},{"model_name":"perplexity/sonar-deep-research","description":"Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics. It autonomously searches, reads, and evaluates sources, refining its approach as it gathers...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"perplexity","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"openrouter_id":"perplexity/sonar-deep-research","context_length":128000}},{"model_name":"perplexity/sonar-reasoning-pro","description":"Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) Sonar Reasoning Pro is a premier reasoning model powered by DeepSeek R1 with Chain of Thought (CoT). Designed for","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"perplexity","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"openrouter_id":"perplexity/sonar-reasoning-pro","context_length":128000}},{"model_name":"qwen/qwen3.8-2.4t-a95b","description":"Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"qwen/qwen3.8-2.4t-a95b","context_length":1048576}},{"model_name":"qwen/qwen3.8-max-0902","description":"Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text,...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"qwen/qwen3.8-max-0902","context_length":1000000}},{"model_name":"Qwen3.8 Max","description":"Qwen3.8 正式旗舰模型，适合复杂编码、专业工作、多模态理解和长程 Agent。","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"Qwen","completion_ratio":3,"cache_ratio":0.125,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Qwen","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"context_length":null}},{"model_name":"qwen3.8-2.4t-a95b","description":"Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"qwen/qwen3.8-2.4t-a95b","context_length":1048576}},{"model_name":"qwen3.8-max-0902","description":"Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text,...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"qwen/qwen3.8-max-0902","context_length":1000000}},{"model_name":"sakana/fugu-max","description":"Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"sakana","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"sakana/fugu-max","context_length":1000000}},{"model_name":"sonar-deep-research","description":"Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics. It autonomously searches, reads, and evaluates sources, refining its approach as it gathers...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"perplexity","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"openrouter_id":"perplexity/sonar-deep-research","context_length":128000}},{"model_name":"sonar-reasoning-pro","description":"Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) Sonar Reasoning Pro is a premier reasoning model powered by DeepSeek R1 with Chain of Thought (CoT). Designed for","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"perplexity","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":2,"output_usd_per_m":8,"openrouter_id":"perplexity/sonar-reasoning-pro","context_length":128000}},{"model_name":"x-ai/grok-4.5","description":"Grok 4.5 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM.","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.15,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"x-ai/grok-4.5","context_length":500000}},{"model_name":"x-ai/grok-4.6","description":"Grok 4.6 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7).","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"x-ai/grok-4.6","context_length":500000}},{"model_name":"x-ai/grok-4.7","description":"Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...","quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"xai","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xai","grokcode":{"input_usd_per_m":2,"output_usd_per_m":6,"openrouter_id":"x-ai/grok-4.7","context_length":500000}},{"model_name":"MiniMax M3","description":"MiniMax 最新 M 系列模型，面向 Agentic reasoning、tool use、coding 和长上下文任务。","quota_type":0,"model_ratio":1.05,"model_price":0,"owner_by":"MiniMax","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,MiniMax","grokcode":{"input_usd_per_m":2.1,"output_usd_per_m":8.4,"context_length":null}},{"model_name":"moonshotai/kimi-k3:batch","description":"Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...","quota_type":0,"model_ratio":1.14,"model_price":0,"owner_by":"kimi","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,kimi","grokcode":{"input_usd_per_m":2.2800000000000002,"output_usd_per_m":11.399999999999999,"openrouter_id":"moonshotai/kimi-k3:batch","context_length":1048576}},{"model_name":"gpt-5.2-pro","description":"gpt-5.2-pro","quota_type":0,"model_ratio":1.1483,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2.29654687,"output_usd_per_m":18.372375,"context_length":null}},{"model_name":"gpt-5.2-pro-2025-12-11","description":"gpt-5.2-pro-2025-12-11","quota_type":0,"model_ratio":1.1483,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":2.29654687,"output_usd_per_m":18.372375,"context_length":null}},{"model_name":"amazon/nova-premier-v1","description":"Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"amazon","completion_ratio":5,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":12.5,"openrouter_id":"amazon/nova-premier-v1","context_length":1000000}},{"model_name":"anthracite-org/magnum-v4-72b","description":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus).\n\nThe model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"anthracite-org","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthracite-org","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":5,"openrouter_id":"anthracite-org/magnum-v4-72b","context_length":32768}},{"model_name":"anthropic/claude-opus-4.5:batch","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":12.5,"openrouter_id":"anthropic/claude-opus-4.5:batch","context_length":200000}},{"model_name":"anthropic/claude-opus-4.6:batch","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":12.5,"openrouter_id":"anthropic/claude-opus-4.6:batch","context_length":1000000}},{"model_name":"anthropic/claude-opus-4.7:batch","description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":12.5,"openrouter_id":"anthropic/claude-opus-4.7:batch","context_length":1000000}},{"model_name":"anthropic/claude-opus-4.8:batch","description":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":12.5,"openrouter_id":"anthropic/claude-opus-4.8:batch","context_length":1000000}},{"model_name":"anthropic/claude-opus-5:batch","description":"Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":12.5,"openrouter_id":"anthropic/claude-opus-5:batch","context_length":1000000}},{"model_name":"Claude Opus 5","description":"Anthropic 旗舰模型，面向长时间运行的 Agent、复杂代码任务、深度推理和高价值专业工作。","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"Claude","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Claude","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":5,"context_length":null}},{"model_name":"cohere/command-a","description":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases. Compared to other leading proprietary...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"cohere/command-a","context_length":256000}},{"model_name":"cohere/command-r-plus-08-2024","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"cohere/command-r-plus-08-2024","context_length":128000}},{"model_name":"command-a","description":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases. Compared to other leading proprietary...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"cohere/command-a","context_length":256000}},{"model_name":"command-r-plus-08-2024","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"cohere","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"cohere/command-r-plus-08-2024","context_length":128000}},{"model_name":"GPT Image 1.5","description":"GPT Image 1.5","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":2.56,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":6.4,"context_length":null}},{"model_name":"gpt-4o","description":"OpenAI 旗舰多模态","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"context_length":null}},{"model_name":"gpt-4o-2024-08-06","description":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format. Read more [here](https://openai.com/index/introducing-structured-outputs-in-the-api/). GPT-4o (\"o\" for \"omni\") is...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"openai/gpt-4o-2024-08-06","context_length":128000}},{"model_name":"gpt-4o-2024-11-20","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"openai/gpt-4o-2024-11-20","context_length":128000}},{"model_name":"gpt-4o-audio-preview","description":"gpt-4o-audio-preview","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"context_length":null}},{"model_name":"gpt-5-image-mini","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":0.8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":2,"openrouter_id":"openai/gpt-5-image-mini","context_length":400000}},{"model_name":"gpt-5.4","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":15,"openrouter_id":"openai/gpt-5.4","context_length":1050000}},{"model_name":"GPT-5.4","description":"OpenAI 高能力通用模型，适合大多数 Coding Agent、检索增强和产品内 AI 功能。","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"OpenAI","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,OpenAI","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":15,"context_length":null}},{"model_name":"gpt-5.4-2026-03-05","description":"gpt-5.4-2026-03-05","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":15,"context_length":null}},{"model_name":"gpt-5.4-openai-compact","description":"gpt-5.4-openai-compact","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":15,"context_length":null}},{"model_name":"gpt-5.5-openai-compact","description":"gpt-5.5-openai-compact","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":2.4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":6,"context_length":null}},{"model_name":"gpt-audio","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"openai/gpt-audio","context_length":128000}},{"model_name":"gpt-reserve","description":"gpt-reserve","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":15,"context_length":null}},{"model_name":"magnum-v4-72b","description":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus).\n\nThe model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"anthracite-org","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthracite-org","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":5,"openrouter_id":"anthracite-org/magnum-v4-72b","context_length":32768}},{"model_name":"nova-premier-v1","description":"Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"amazon","completion_ratio":5,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,amazon","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":12.5,"openrouter_id":"amazon/nova-premier-v1","context_length":1000000}},{"model_name":"openai/gpt-4o","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"openai/gpt-4o","context_length":128000}},{"model_name":"openai/gpt-4o-2024-08-06","description":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format. Read more [here](https://openai.com/index/introducing-structured-outputs-in-the-api/). GPT-4o (\"o\" for \"omni\") is...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"openai/gpt-4o-2024-08-06","context_length":128000}},{"model_name":"openai/gpt-4o-2024-11-20","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"openai/gpt-4o-2024-11-20","context_length":128000}},{"model_name":"openai/gpt-5-image-mini","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":0.8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":2,"openrouter_id":"openai/gpt-5-image-mini","context_length":400000}},{"model_name":"openai/gpt-5.4","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":15,"openrouter_id":"openai/gpt-5.4","context_length":1050000}},{"model_name":"openai/gpt-5.5:batch","description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":15,"openrouter_id":"openai/gpt-5.5:batch","context_length":1050000}},{"model_name":"openai/gpt-audio","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...","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":10,"openrouter_id":"openai/gpt-audio","context_length":128000}},{"model_name":"pareto-26.9","description":"Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"unbiased","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,unbiased","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":7.5,"openrouter_id":"unbiased/pareto-26.9","context_length":262144}},{"model_name":"Qwen3.7 Max","description":"Qwen 当前高能力模型之一，适合复杂代码与 Agent 任务，已补齐阿里云百炼官方按量价格。","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"Qwen","completion_ratio":3,"cache_ratio":0.2,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Qwen","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":7.5,"context_length":null}},{"model_name":"unbiased/pareto-26.9","description":"Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.","quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"unbiased","completion_ratio":3,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,unbiased","grokcode":{"input_usd_per_m":2.5,"output_usd_per_m":7.5,"openrouter_id":"unbiased/pareto-26.9","context_length":262144}},{"model_name":"glm-5.3-prime","description":"GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token...","quota_type":0,"model_ratio":1.4,"model_price":0,"owner_by":"glm","completion_ratio":3.1429,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":2.8,"output_usd_per_m":8.8,"openrouter_id":"z-ai/glm-5.3-prime","context_length":1000000}},{"model_name":"z-ai/glm-5.3-prime","description":"GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token...","quota_type":0,"model_ratio":1.4,"model_price":0,"owner_by":"glm","completion_ratio":3.1429,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,glm","grokcode":{"input_usd_per_m":2.8,"output_usd_per_m":8.8,"openrouter_id":"z-ai/glm-5.3-prime","context_length":1000000}},{"model_name":"aion-3.0","description":"Aion-3.0 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each contribute...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":3,"output_usd_per_m":6,"openrouter_id":"aion-labs/aion-3.0","context_length":131072}},{"model_name":"aion-3.5","description":"Aion 3.5 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":3,"output_usd_per_m":6,"openrouter_id":"aion-labs/aion-3.5","context_length":262144}},{"model_name":"aion-labs/aion-3.0","description":"Aion-3.0 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each contribute...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":3,"output_usd_per_m":6,"openrouter_id":"aion-labs/aion-3.0","context_length":131072}},{"model_name":"aion-labs/aion-3.5","description":"Aion 3.5 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"aion-labs","completion_ratio":2,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,aion-labs","grokcode":{"input_usd_per_m":3,"output_usd_per_m":6,"openrouter_id":"aion-labs/aion-3.5","context_length":262144}},{"model_name":"anthropic/claude-sonnet-4.5","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"anthropic/claude-sonnet-4.5","context_length":1000000}},{"model_name":"anthropic/claude-sonnet-4.6","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"anthropic/claude-sonnet-4.6","context_length":1000000}},{"model_name":"Claude Sonnet 4.6","description":"Anthropic Claude Sonnet 系列均衡模型，适合日常开发、长上下文阅读和 Agent 工作流。","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"Claude","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,Claude","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"context_length":null}},{"model_name":"claude-3-5-sonnet-20240620","description":"claude-3-5-sonnet-20240620","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"context_length":null}},{"model_name":"claude-3-5-sonnet-20241022","description":"claude-3-5-sonnet-20241022","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"context_length":null}},{"model_name":"claude-3-7-sonnet-20250219","description":"claude-3-7-sonnet-20250219","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"context_length":null}},{"model_name":"claude-sonnet-4","description":"编码与长上下文","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"claude","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,claude","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"context_length":null}},{"model_name":"claude-sonnet-4-20250514","description":"claude-sonnet-4-20250514","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"context_length":null}},{"model_name":"claude-sonnet-4-5-20250929","description":"claude-sonnet-4-5-20250929","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"context_length":null}},{"model_name":"claude-sonnet-4.5","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"anthropic/claude-sonnet-4.5","context_length":1000000}},{"model_name":"claude-sonnet-4.6","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"anthropic/claude-sonnet-4.6","context_length":1000000}},{"model_name":"ember-1","description":"Ember-1 is a specialized reasoning model from Fireworks Research, built on [Kimi K3](https://openrouter.ai/moonshotai/kimi-k3). It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40%...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"fireworks","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,fireworks","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"fireworks/ember-1","context_length":1048576}},{"model_name":"fireworks/ember-1","description":"Ember-1 is a specialized reasoning model from Fireworks Research, built on [Kimi K3](https://openrouter.ai/moonshotai/kimi-k3). It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40%...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"fireworks","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,fireworks","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"fireworks/ember-1","context_length":1048576}},{"model_name":"gpt-3.5-turbo-16k","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"openai","completion_ratio":1.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":3,"output_usd_per_m":4,"openrouter_id":"openai/gpt-3.5-turbo-16k","context_length":16385}},{"model_name":"openai/gpt-3.5-turbo-16k","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"openai","completion_ratio":1.3333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":3,"output_usd_per_m":4,"openrouter_id":"openai/gpt-3.5-turbo-16k","context_length":16385}},{"model_name":"perplexity/sonar-pro","description":"Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) For enterprises seeking more advanced capabilities, the Sonar Pro API can handle in-depth, multi-step queries wit","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"perplexity","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"perplexity/sonar-pro","context_length":200000}},{"model_name":"perplexity/sonar-pro-search","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"perplexity","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"perplexity/sonar-pro-search","context_length":200000}},{"model_name":"sonar-pro","description":"Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) For enterprises seeking more advanced capabilities, the Sonar Pro API can handle in-depth, multi-step queries wit","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"perplexity","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"perplexity/sonar-pro","context_length":200000}},{"model_name":"sonar-pro-search","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...","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"perplexity","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,perplexity","grokcode":{"input_usd_per_m":3,"output_usd_per_m":15,"openrouter_id":"perplexity/sonar-pro-search","context_length":200000}},{"model_name":"v4-flash","description":"DeepSeek V4 的低成本高吞吐版本，适合高频编码、总结、抽取和日常 Agent 调用。","quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"DeepSeek","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,DeepSeek","grokcode":{"input_usd_per_m":3,"output_usd_per_m":9,"context_length":null}},{"model_name":"claude-opus-4-1-20250805","description":"claude-opus-4-1-20250805","quota_type":0,"model_ratio":1.6667,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,unknown","grokcode":{"input_usd_per_m":3.33333333,"output_usd_per_m":16.66666667,"context_length":null}},{"model_name":"claude-opus-4-20250514","description":"claude-opus-4-20250514","quota_type":0,"model_ratio":1.6667,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,unknown","grokcode":{"input_usd_per_m":3.33333333,"output_usd_per_m":16.66666667,"context_length":null}},{"model_name":"~anthropic/claude-opus-latest","description":"This model always redirects to the latest model in the Claude Opus family.","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"~anthropic/claude-opus-latest","context_length":1000000}},{"model_name":"anthropic/claude-opus-5.5","description":"Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"anthropic/claude-opus-5.5","context_length":1000000}},{"model_name":"claude-opus-5.5","description":"Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"anthropic/claude-opus-5.5","context_length":1000000}},{"model_name":"claude-opus-latest","description":"This model always redirects to the latest model in the Claude Opus family.","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.05,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"~anthropic/claude-opus-latest","context_length":1000000}},{"model_name":"gpt-5.6-sol","description":"GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"openai/gpt-5.6-sol","context_length":1050000}},{"model_name":"gpt-5.6-sol-pro","description":"GPT-5.6 Sol Pro is the same underlying model as [GPT-5.6 Sol](https://openrouter.ai/openai/gpt-5.6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"openai/gpt-5.6-sol-pro","context_length":1050000}},{"model_name":"meta/muse-wonder","description":"Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"llama","completion_ratio":5,"cache_ratio":0.05,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"meta/muse-wonder","context_length":1048576}},{"model_name":"muse-wonder","description":"Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"llama","completion_ratio":5,"cache_ratio":0.05,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,llama","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"meta/muse-wonder","context_length":1048576}},{"model_name":"openai/gpt-5.6-sol","description":"GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"openai/gpt-5.6-sol","context_length":1050000}},{"model_name":"openai/gpt-5.6-sol-pro","description":"GPT-5.6 Sol Pro is the same underlying model as [GPT-5.6 Sol](https://openrouter.ai/openai/gpt-5.6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":4,"output_usd_per_m":20,"openrouter_id":"openai/gpt-5.6-sol-pro","context_length":1050000}},{"model_name":"qwen/qwen3.8-max-prime","description":"Qwen3.8 Max Prime is a higher-throughput variant of Qwen3.8 Max from Alibaba's Qwen team, served as a separate SKU at a higher price point. It accepts text, image, and video...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":4,"output_usd_per_m":12,"openrouter_id":"qwen/qwen3.8-max-prime","context_length":1000000}},{"model_name":"qwen3.8-max-prime","description":"Qwen3.8 Max Prime is a higher-throughput variant of Qwen3.8 Max from Alibaba's Qwen team, served as a separate SKU at a higher price point. It accepts text, image, and video...","quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"qwen","completion_ratio":3,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,qwen","grokcode":{"input_usd_per_m":4,"output_usd_per_m":12,"openrouter_id":"qwen/qwen3.8-max-prime","context_length":1000000}},{"model_name":"mimo-v2.6-pro-ultraspeed","description":"MiMo-V2.6-Pro-UltraSpeed is the fast speed edition of Xiaomi's flagship foundation model, MiMo-V2.6-Pro. Built from the same 1T MiMo-V2.6-Pro checkpoint, it matches the original model in quality while delivering roughly 10x...","quota_type":0,"model_ratio":2.175,"model_price":0,"owner_by":"xiaomi","completion_ratio":2,"cache_ratio":0.008276,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":4.35,"output_usd_per_m":8.7,"openrouter_id":"xiaomi/mimo-v2.6-pro-ultraspeed","context_length":1048576}},{"model_name":"xiaomi/mimo-v2.6-pro-ultraspeed","description":"MiMo-V2.6-Pro-UltraSpeed is the fast speed edition of Xiaomi's flagship foundation model, MiMo-V2.6-Pro. Built from the same 1T MiMo-V2.6-Pro checkpoint, it matches the original model in quality while delivering roughly 10x...","quota_type":0,"model_ratio":2.175,"model_price":0,"owner_by":"xiaomi","completion_ratio":2,"cache_ratio":0.008276,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xiaomi","grokcode":{"input_usd_per_m":4.35,"output_usd_per_m":8.7,"openrouter_id":"xiaomi/mimo-v2.6-pro-ultraspeed","context_length":1048576}},{"model_name":"v4-pro","description":"DeepSeek V4 的旗舰能力版本，适合更复杂的推理、长上下文和高失败成本任务。","quota_type":0,"model_ratio":2.25,"model_price":0,"owner_by":"DeepSeek","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,DeepSeek","grokcode":{"input_usd_per_m":4.5,"output_usd_per_m":13.5,"context_length":null}},{"model_name":"anthropic/claude-fable-5:batch","description":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-fable-5:batch","context_length":1000000}},{"model_name":"anthropic/claude-fable-5.1:batch","description":"Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.025,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-fable-5.1:batch","context_length":1000000}},{"model_name":"anthropic/claude-opus-4.5","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...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.5","context_length":200000}},{"model_name":"anthropic/claude-opus-4.6","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...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.6","context_length":1000000}},{"model_name":"anthropic/claude-opus-4.7","description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.7","context_length":1000000}},{"model_name":"anthropic/claude-opus-4.8","description":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.8","context_length":1000000}},{"model_name":"anthropic/claude-opus-5","description":"Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-5","context_length":1000000}},{"model_name":"Claude Fable 5","description":"Anthropic Claude 前沿高能力模型，适合复杂推理、深度代码库任务和高价值 Agent 工作流。","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"Claude","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,Claude","grokcode":{"input_usd_per_m":5,"output_usd_per_m":5,"context_length":null}},{"model_name":"Claude Opus 4.6","description":"Anthropic Claude Opus 系列高能力模型，适合复杂推理、长任务编排和高价值编码场景。","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"Claude","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,Claude","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"context_length":null}},{"model_name":"Claude Opus 4.7","description":"Anthropic Claude Opus 系列高能力模型，适合复杂代码库理解、深度推理和高价值 Agent 任务。","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"Claude","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,Claude","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"context_length":null}},{"model_name":"Claude Opus 4.8","description":"Anthropic Claude 高能力模型，适合复杂代码库理解、深度推理和高价值 Agent 任务。","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"Claude","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,Claude","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"context_length":null}},{"model_name":"claude-opus-4-5-20251101","description":"claude-opus-4-5-20251101","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"unknown","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,unknown","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"context_length":null}},{"model_name":"claude-opus-4.5","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...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.5","context_length":200000}},{"model_name":"claude-opus-4.6","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...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.6","context_length":1000000}},{"model_name":"claude-opus-4.7","description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.7","context_length":1000000}},{"model_name":"claude-opus-4.8","description":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-4.8","context_length":1000000}},{"model_name":"claude-opus-5","description":"Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"anthropic/claude-opus-5","context_length":1000000}},{"model_name":"Codex Auto Review","description":"Codex Auto Review","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"context_length":null}},{"model_name":"fugu-ultra","description":"Fugu Ultra is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"sakana","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"sakana/fugu-ultra","context_length":1000000}},{"model_name":"fugu-ultra-v2","description":"Fugu Ultra v2 is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"sakana","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"sakana/fugu-ultra-v2","context_length":1000000}},{"model_name":"GPT Image 2","description":"OpenAI 图片生成标准模型，沿用中转 API 图片生成分组，主要作为官方输入、缓存和图片输出价格基准。","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"图片生成","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,图片生成","grokcode":{"input_usd_per_m":5,"output_usd_per_m":10,"context_length":null}},{"model_name":"gpt-4-turbo","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.","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":15,"openrouter_id":"openai/gpt-4-turbo:batch","context_length":128000}},{"model_name":"gpt-4o-2024-05-13","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...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":15,"openrouter_id":"openai/gpt-4o-2024-05-13","context_length":128000}},{"model_name":"gpt-5.5","description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"openai/gpt-5.5","context_length":1050000}},{"model_name":"GPT-5.5","description":"OpenAI 当前旗舰文本与 Agent 模型，适合高失败成本的复杂推理、代码审查和长任务编排。","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"OpenAI","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,OpenAI","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"context_length":null}},{"model_name":"GPT-5.6 (Sol)","description":"GPT-5.6 (Sol)","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"context_length":null}},{"model_name":"GPT-5.6 Sol","description":"GPT-5.6 Sol","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"context_length":null}},{"model_name":"gpt-chat-latest","description":"GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT. As OpenAI rolls out new Instant model updates...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"openai/gpt-chat-latest","context_length":400000}},{"model_name":"openai/gpt-4-turbo:batch","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.","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":15,"openrouter_id":"openai/gpt-4-turbo:batch","context_length":128000}},{"model_name":"openai/gpt-4o-2024-05-13","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...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":15,"openrouter_id":"openai/gpt-4o-2024-05-13","context_length":128000}},{"model_name":"openai/gpt-5.5","description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"openai/gpt-5.5","context_length":1050000}},{"model_name":"openai/gpt-6-astra-pro:batch","description":"GPT-6 Astra Pro is the same underlying model as [GPT-6 Astra](https://openrouter.ai/openai/gpt-6-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"openai/gpt-6-astra-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-6-astra:batch","description":"GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":25,"openrouter_id":"openai/gpt-6-astra:batch","context_length":1050000}},{"model_name":"openai/gpt-chat-latest","description":"GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT. As OpenAI rolls out new Instant model updates...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"openai/gpt-chat-latest","context_length":400000}},{"model_name":"sakana/fugu-ultra","description":"Fugu Ultra is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"sakana","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"sakana/fugu-ultra","context_length":1000000}},{"model_name":"sakana/fugu-ultra-v2","description":"Fugu Ultra v2 is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to...","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"sakana","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,sakana","grokcode":{"input_usd_per_m":5,"output_usd_per_m":30,"openrouter_id":"sakana/fugu-ultra-v2","context_length":1000000}},{"model_name":"anthropic/claude-opus-4.1:batch","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...","quota_type":0,"model_ratio":3.75,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":7.5,"output_usd_per_m":37.5,"openrouter_id":"anthropic/claude-opus-4.1:batch","context_length":200000}},{"model_name":"openai/gpt-5-pro:batch","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...","quota_type":0,"model_ratio":3.75,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":7.5,"output_usd_per_m":60,"openrouter_id":"openai/gpt-5-pro:batch","context_length":400000}},{"model_name":"gpt-5.4-image-2","description":"[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2. It enables rich multimodal workflows, allowing users to seamlessly move between reasoning, coding, and...","quota_type":0,"model_ratio":4,"model_price":0,"owner_by":"openai","completion_ratio":1.875,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":8,"output_usd_per_m":15,"openrouter_id":"openai/gpt-5.4-image-2","context_length":272000}},{"model_name":"openai/gpt-5.4-image-2","description":"[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2. It enables rich multimodal workflows, allowing users to seamlessly move between reasoning, coding, and...","quota_type":0,"model_ratio":4,"model_price":0,"owner_by":"openai","completion_ratio":1.875,"cache_ratio":0.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":8,"output_usd_per_m":15,"openrouter_id":"openai/gpt-5.4-image-2","context_length":272000}},{"model_name":"~anthropic/claude-fable-latest","description":"This model always redirects to the latest model in the Claude Fable family.","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.025,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"~anthropic/claude-fable-latest","context_length":1000000}},{"model_name":"~openai/gpt-astra-latest","description":"This model always redirects to the latest model in the GPT Astra family.","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"~openai/gpt-astra-latest","context_length":1050000}},{"model_name":"anthropic/claude-fable-5","description":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"anthropic/claude-fable-5","context_length":1000000}},{"model_name":"anthropic/claude-fable-5.1","description":"Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.025,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"anthropic/claude-fable-5.1","context_length":1000000}},{"model_name":"claude-fable-5","description":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"anthropic/claude-fable-5","context_length":1000000}},{"model_name":"claude-fable-5.1","description":"Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.025,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"anthropic/claude-fable-5.1","context_length":1000000}},{"model_name":"claude-fable-latest","description":"This model always redirects to the latest model in the Claude Fable family.","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"~anthropic","completion_ratio":5,"cache_ratio":0.025,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~anthropic","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"~anthropic/claude-fable-latest","context_length":1000000}},{"model_name":"gpt-5-image","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,...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"openai","completion_ratio":1,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5-image","context_length":400000}},{"model_name":"gpt-6-astra","description":"GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"openai/gpt-6-astra","context_length":1050000}},{"model_name":"gpt-6-astra-pro","description":"GPT-6 Astra Pro is the same underlying model as [GPT-6 Astra](https://openrouter.ai/openai/gpt-6-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"openai/gpt-6-astra-pro","context_length":1050000}},{"model_name":"gpt-astra-latest","description":"This model always redirects to the latest model in the GPT Astra family.","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"~openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,~openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"~openai/gpt-astra-latest","context_length":1050000}},{"model_name":"openai/gpt-5-image","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,...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"openai","completion_ratio":1,"cache_ratio":0.125,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":10,"openrouter_id":"openai/gpt-5-image","context_length":400000}},{"model_name":"openai/gpt-6-astra","description":"GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"openai/gpt-6-astra","context_length":1050000}},{"model_name":"openai/gpt-6-astra-pro","description":"GPT-6 Astra Pro is the same underlying model as [GPT-6 Astra](https://openrouter.ai/openai/gpt-6-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...","quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"openai","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":10,"output_usd_per_m":50,"openrouter_id":"openai/gpt-6-astra-pro","context_length":1050000}},{"model_name":"openai/gpt-5.2-pro:batch","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,...","quota_type":0,"model_ratio":5.25,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":10.5,"output_usd_per_m":84,"openrouter_id":"openai/gpt-5.2-pro:batch","context_length":400000}},{"model_name":"anthropic/claude-opus-4.1","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...","quota_type":0,"model_ratio":7.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":15,"output_usd_per_m":75,"openrouter_id":"anthropic/claude-opus-4.1","context_length":200000}},{"model_name":"claude-opus-4.1","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...","quota_type":0,"model_ratio":7.5,"model_price":0,"owner_by":"anthropic","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,anthropic","grokcode":{"input_usd_per_m":15,"output_usd_per_m":75,"openrouter_id":"anthropic/claude-opus-4.1","context_length":200000}},{"model_name":"gpt-5-pro","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...","quota_type":0,"model_ratio":7.5,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":15,"output_usd_per_m":120,"openrouter_id":"openai/gpt-5-pro","context_length":400000}},{"model_name":"openai/gpt-5-pro","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...","quota_type":0,"model_ratio":7.5,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":15,"output_usd_per_m":120,"openrouter_id":"openai/gpt-5-pro","context_length":400000}},{"model_name":"openai/gpt-5.4-pro:batch","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...","quota_type":0,"model_ratio":7.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":15,"output_usd_per_m":90,"openrouter_id":"openai/gpt-5.4-pro:batch","context_length":1050000}},{"model_name":"openai/gpt-5.5-pro:batch","description":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads. It features a 1M+ token context window (922K input, 128K output) with support for...","quota_type":0,"model_ratio":7.5,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":15,"output_usd_per_m":90,"openrouter_id":"openai/gpt-5.5-pro:batch","context_length":1050000}},{"model_name":"o3-pro","description":"The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently...","quota_type":0,"model_ratio":10,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":20,"output_usd_per_m":80,"openrouter_id":"openai/o3-pro","context_length":200000}},{"model_name":"openai/o3-pro","description":"The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently...","quota_type":0,"model_ratio":10,"model_price":0,"owner_by":"openai","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":20,"output_usd_per_m":80,"openrouter_id":"openai/o3-pro","context_length":200000}},{"model_name":"openai/gpt-5.2-pro","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,...","quota_type":0,"model_ratio":10.5,"model_price":0,"owner_by":"openai","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":21,"output_usd_per_m":168,"openrouter_id":"openai/gpt-5.2-pro","context_length":400000}},{"model_name":"gpt-4","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...","quota_type":0,"model_ratio":15,"model_price":0,"owner_by":"openai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":30,"output_usd_per_m":60,"openrouter_id":"openai/gpt-4","context_length":8191}},{"model_name":"gpt-5.4-pro","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...","quota_type":0,"model_ratio":15,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":30,"output_usd_per_m":180,"openrouter_id":"openai/gpt-5.4-pro","context_length":1050000}},{"model_name":"gpt-5.5-pro","description":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads. It features a 1M+ token context window (922K input, 128K output) with support for...","quota_type":0,"model_ratio":15,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":30,"output_usd_per_m":180,"openrouter_id":"openai/gpt-5.5-pro","context_length":1050000}},{"model_name":"openai/gpt-4","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...","quota_type":0,"model_ratio":15,"model_price":0,"owner_by":"openai","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":30,"output_usd_per_m":60,"openrouter_id":"openai/gpt-4","context_length":8191}},{"model_name":"openai/gpt-5.4-pro","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...","quota_type":0,"model_ratio":15,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":30,"output_usd_per_m":180,"openrouter_id":"openai/gpt-5.4-pro","context_length":1050000}},{"model_name":"openai/gpt-5.5-pro","description":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads. It features a 1M+ token context window (922K input, 128K output) with support for...","quota_type":0,"model_ratio":15,"model_price":0,"owner_by":"openai","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openai","grokcode":{"input_usd_per_m":30,"output_usd_per_m":180,"openrouter_id":"openai/gpt-5.5-pro","context_length":1050000}},{"model_name":"cohere/north-mini-code:free","description":"North Mini Code is Cohere's first agentic coding model and the debut of its North family. A sparse mixture-of-experts model with 30B total parameters and 3B active, it is optimized...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"cohere","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"cohere/north-mini-code:free","context_length":256000}},{"model_name":"crouton-alpha","description":"Crouton Alpha is a lightweight model built for scalable coding, agentic, and multimodal workflows. It's ideal for executing agentic sub-tasks, tool use, and document analysis in high-volume, repeated work. Developers...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"stealth","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stealth","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"stealth/crouton-alpha","context_length":1048576}},{"model_name":"dots-3-note-preview","description":"Dots3-Note Preview is an open-weight mixture-of-experts model from Dots Studio, with 16B active parameters out of 280B total. It is the lightest model in the Dots 3 family and is...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"dots-studio","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,dots-studio","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"dots-studio/dots-3-note-preview:free","context_length":512000}},{"model_name":"dots-studio/dots-3-note-preview:free","description":"Dots3-Note Preview is an open-weight mixture-of-experts model from Dots Studio, with 16B active parameters out of 280B total. It is the lightest model in the Dots 3 family and is...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"dots-studio","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,dots-studio","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"dots-studio/dots-3-note-preview:free","context_length":512000}},{"model_name":"flint-v1.0","description":"Flint is an open-weight mixture-of-experts model from Recursive, with about 2.5B active parameters out of 40B total, built by Recursive's autoresearch system Eureka Machine for coding agents. It is suited...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"recursive","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,recursive","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"recursive/flint-v1.0:free","context_length":262144}},{"model_name":"free","description":"The simplest way to get free inference. openrouter/free is a router that selects free models at random from the models available on OpenRouter. The router smartly filters for models that...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"openrouter","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openrouter","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"openrouter/free","context_length":200000}},{"model_name":"google/gemma-4-26b-a4b-it:free","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...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"gemma","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"google/gemma-4-26b-a4b-it:free","context_length":262144}},{"model_name":"google/gemma-4-31b-it:free","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...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"gemma","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,gemma","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"google/gemma-4-31b-it:free","context_length":262144}},{"model_name":"google/lyria-3-clip-preview","description":"30 second duration clips are priced at $0.04 per clip. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"google","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"google/lyria-3-clip-preview","context_length":1048576}},{"model_name":"google/lyria-3-pro-preview","description":"Full-length songs are priced at $0.08 per song. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate high-quality, 48kHz...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"google","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"google/lyria-3-pro-preview","context_length":1048576}},{"model_name":"GPT Image 2.5 Flare","description":"GPT Image 2.5 Flare","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"openai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"GPT Image 2.5 Sunburst","description":"GPT Image 2.5 Sunburst","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"openai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,openai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"gpt-image-2.5","description":"gpt-image-2.5","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"openai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,openai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"Grok Imagine Image","description":"Grok Imagine Image","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"xai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,xai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"Grok Imagine Image 2.0","description":"Grok Imagine Image 2.0","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"xai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,xai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"Grok Imagine Image Quality","description":"Grok Imagine Image Quality","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"xai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"official,xai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"Grok Imagine Video","description":"Grok Imagine Video","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"xai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"Grok Imagine Video 1.5","description":"Grok Imagine Video 1.5","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"xai","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"merged,xai","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"context_length":null}},{"model_name":"liquid/lfm-2.5-2.6b:free","description":"LFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"liquid","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,liquid","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"liquid/lfm-2.5-2.6b:free","context_length":65536}},{"model_name":"lyria-3-clip-preview","description":"30 second duration clips are priced at $0.04 per clip. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"google","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"google/lyria-3-clip-preview","context_length":1048576}},{"model_name":"lyria-3-pro-preview","description":"Full-length songs are priced at $0.08 per song. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate high-quality, 48kHz...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"google","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,google","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"google/lyria-3-pro-preview","context_length":1048576}},{"model_name":"nemotron-3-nano-omni-30b-a3b-reasoning","description":"NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems. It accepts text, image, video, and...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free","context_length":256000}},{"model_name":"north-mini-code","description":"North Mini Code is Cohere's first agentic coding model and the debut of its North family. A sparse mixture-of-experts model with 30B total parameters and 3B active, it is optimized...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"cohere","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,cohere","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"cohere/north-mini-code:free","context_length":256000}},{"model_name":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free","description":"NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems. It accepts text, image, video, and...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free","context_length":256000}},{"model_name":"nvidia/nemotron-3-ultra-550b-a55b:free","description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"nvidia/nemotron-3-ultra-550b-a55b:free","context_length":1000000}},{"model_name":"nvidia/nemotron-3.5-content-safety:free","description":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It moderates both inputs to and responses from LLMs and VLMs, accepting...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"nvidia/nemotron-3.5-content-safety:free","context_length":128000}},{"model_name":"nvidia/nemotron-3.5-lightning:free","description":"NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"nvidia/nemotron-3.5-lightning:free","context_length":1000000}},{"model_name":"nvidia/nemotron-3.5-super-vl:free","description":"NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"nvidia","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,nvidia","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"nvidia/nemotron-3.5-super-vl:free","context_length":1000000}},{"model_name":"openrouter/free","description":"The simplest way to get free inference. openrouter/free is a router that selects free models at random from the models available on OpenRouter. The router smartly filters for models that...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"openrouter","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,openrouter","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"openrouter/free","context_length":200000}},{"model_name":"recursive/flint-v1.0:free","description":"Flint is an open-weight mixture-of-experts model from Recursive, with about 2.5B active parameters out of 40B total, built by Recursive's autoresearch system Eureka Machine for coding agents. It is suited...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"recursive","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,recursive","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"recursive/flint-v1.0:free","context_length":262144}},{"model_name":"stealth/crouton-alpha","description":"Crouton Alpha is a lightweight model built for scalable coding, agentic, and multimodal workflows. It's ideal for executing agentic sub-tasks, tool use, and document analysis in high-volume, repeated work. Developers...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"stealth","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,stealth","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"stealth/crouton-alpha","context_length":1048576}},{"model_name":"xing4.0-29b-a4b","description":"Xing4.0-29B-A4B is an open-weight mixture-of-experts language model from China Telecom's AI team (TeleAI) and the successor to the TeleChat series. It has 29B total parameters with 4B active per token...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"xingchen-agi","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xingchen-agi","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"xingchen-agi/xing4.0-29b-a4b:free","context_length":262144}},{"model_name":"xingchen-agi/xing4.0-29b-a4b:free","description":"Xing4.0-29B-A4B is an open-weight mixture-of-experts language model from China Telecom's AI team (TeleAI) and the successor to the TeleChat series. It has 29B total parameters with 4B active per token...","quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"xingchen-agi","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"billing_mode":"token","tags":"openrouter,xingchen-agi","grokcode":{"input_usd_per_m":0,"output_usd_per_m":0,"openrouter_id":"xingchen-agi/xing4.0-29b-a4b:free","context_length":262144}}],"vendors":[{"id":1,"name":"openai","description":"openai models"},{"id":2,"name":"视频生成","description":"视频生成 models"},{"id":3,"name":"~deepseek","description":"~deepseek models"},{"id":4,"name":"glm","description":"glm models"},{"id":5,"name":"inclusionai","description":"inclusionai models"},{"id":6,"name":"nex-agi","description":"nex-agi models"},{"id":7,"name":"llama","description":"llama models"},{"id":8,"name":"~z-ai","description":"~z-ai models"},{"id":9,"name":"mistral","description":"mistral models"},{"id":10,"name":"inference-net","description":"inference-net models"},{"id":11,"name":"qwen","description":"qwen models"},{"id":12,"name":"amazon","description":"amazon models"},{"id":13,"name":"cohere","description":"cohere models"},{"id":14,"name":"scaledown","description":"scaledown models"},{"id":15,"name":"inception","description":"inception models"},{"id":16,"name":"sao10k","description":"sao10k models"},{"id":17,"name":"deepseek","description":"deepseek models"},{"id":18,"name":"tencent","description":"tencent models"},{"id":19,"name":"anthropic","description":"anthropic models"},{"id":20,"name":"gemma","description":"gemma models"},{"id":21,"name":"liquid","description":"liquid models"},{"id":22,"name":"particle-ai","description":"particle-ai models"},{"id":23,"name":"upstage","description":"upstage models"},{"id":24,"name":"ibm-granite","description":"ibm-granite models"},{"id":25,"name":"nvidia","description":"nvidia models"},{"id":26,"name":"microsoft","description":"microsoft models"},{"id":27,"name":"bytedance-seed","description":"bytedance-seed models"},{"id":28,"name":"GLM","description":"GLM models"},{"id":29,"name":"minimax","description":"minimax models"},{"id":30,"name":"prism-ml","description":"prism-ml models"},{"id":31,"name":"gryphe","description":"gryphe models"},{"id":32,"name":"~anthropic","description":"~anthropic models"},{"id":33,"name":"~openai","description":"~openai models"},{"id":34,"name":"bytedance","description":"bytedance models"},{"id":35,"name":"google","description":"google models"},{"id":36,"name":"rekaai","description":"rekaai models"},{"id":37,"name":"stepfun","description":"stepfun models"},{"id":38,"name":"MiMo","description":"MiMo models"},{"id":39,"name":"xiaomi","description":"xiaomi models"},{"id":40,"name":"perceptron","description":"perceptron models"},{"id":41,"name":"Qwen","description":"Qwen models"},{"id":42,"name":"~moonshotai","description":"~moonshotai models"},{"id":43,"name":"kimi","description":"kimi models"},{"id":44,"name":"cognitivecomputations","description":"cognitivecomputations models"},{"id":45,"name":"ricursiveai","description":"ricursiveai models"},{"id":46,"name":"OpenAI","description":"OpenAI models"},{"id":47,"name":"Gemini","description":"Gemini models"},{"id":48,"name":"图片生成","description":"图片生成 models"},{"id":49,"name":"reflection","description":"reflection models"},{"id":50,"name":"thedrummer","description":"thedrummer models"},{"id":51,"name":"meituan","description":"meituan models"},{"id":52,"name":"MiniMax","description":"MiniMax models"},{"id":53,"name":"moondream","description":"moondream models"},{"id":54,"name":"undi95","description":"undi95 models"},{"id":55,"name":"mancer","description":"mancer models"},{"id":56,"name":"thinkingmachines","description":"thinkingmachines models"},{"id":57,"name":"elorian","description":"elorian models"},{"id":58,"name":"Kimi","description":"Kimi models"},{"id":59,"name":"writer","description":"writer models"},{"id":60,"name":"aion-labs","description":"aion-labs models"},{"id":61,"name":"~google","description":"~google models"},{"id":62,"name":"~unbiased","description":"~unbiased models"},{"id":63,"name":"morph","description":"morph models"},{"id":64,"name":"unbiased","description":"unbiased models"},{"id":65,"name":"unknown","description":"unknown models"},{"id":66,"name":"relace","description":"relace models"},{"id":67,"name":"sakana","description":"sakana models"},{"id":68,"name":"Claude","description":"Claude models"},{"id":69,"name":"xai","description":"xai models"},{"id":70,"name":"nousresearch","description":"nousresearch models"},{"id":71,"name":"perplexity","description":"perplexity models"},{"id":72,"name":"gemini","description":"gemini models"},{"id":73,"name":"Grok","description":"Grok models"},{"id":74,"name":"~x-ai","description":"~x-ai models"},{"id":75,"name":"anthracite-org","description":"anthracite-org models"},{"id":76,"name":"claude","description":"claude models"},{"id":77,"name":"fireworks","description":"fireworks models"},{"id":78,"name":"DeepSeek","description":"DeepSeek models"},{"id":79,"name":"stealth","description":"stealth models"},{"id":80,"name":"dots-studio","description":"dots-studio models"},{"id":81,"name":"recursive","description":"recursive models"},{"id":82,"name":"openrouter","description":"openrouter models"},{"id":83,"name":"xingchen-agi","description":"xingchen-agi models"}],"group_ratio":{"default":1},"usable_group":{"default":"default"},"auto_groups":["default"],"supported_endpoint":{"openai":{"path":"/v1/chat/completions","method":"POST"}},"pricing_version":"grokcode-2026-11-05","grokcode_meta":{"source":"grokcode:merged","degraded":false,"message":"合并：官方 380 + OpenRouter 模型 454 → 912 条","fetchedAt":"2026-11-05T23:17:21.592Z","official_count":380,"openrouter_count":454,"total":912,"formula":"model_ratio = input_usd_per_m / 2 ; completion_ratio = output / input ; 1 ratio = $2/M input","homepage":"https://www.grokcode.cn/price-api"}}