{"schemaVersion":1,"referenceBriefs":{"deepseek/deepseek-v4-pro-0813":{"id":"deepseek/deepseek-v4-pro-0813","name":"DeepSeek-V4-Pro-0813","facts":{"family":{"value":"DeepSeek V4","status":"verified","sources":[{"label":"DeepSeek-V4-Pro-0813","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813"}]},"developer":{"value":"DeepSeek-AI","status":"verified","sources":[{"label":"DeepSeek-V4-Pro-0813","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"DeepSeek Privacy Policy","url":"https://cdn.deepseek.com/policies/en-US/deepseek-privacy-policy.html"}]},"releasedAt":{"value":"2026-08-13","status":"verified","sources":[{"label":"DeepSeek API change log","url":"https://api-docs.deepseek.com/updates"}]},"license":{"value":"MIT","status":"verified","sources":[{"label":"DeepSeek-V4-Pro-0813 LICENSE","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/blob/main/LICENSE"}],"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/blob/main/LICENSE"},"architecture":{"value":"Mixture-of-Experts with DeepSeek V4 architecture and DSpark speculative-decoding module","status":"verified","sources":[{"label":"DeepSeek-V4-Pro-0813","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813"},{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]},"totalParameters":{"value":1600000000000,"status":"verified","sources":[{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]},"maxOutputTokens":{"value":393216,"status":"verified","sources":[{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"DeepSeek-V4-Pro-0813 LICENSE","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/blob/main/LICENSE"}]},"activeParameters":{"value":49000000000,"status":"verified","sources":[{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/deepseek/deepseek-v4-pro-0813","providerModelId":"deepseek/deepseek-v4-pro-0813"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"deepseek/deepseek-v4-pro-0813"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813","providerModelId":"deepseek-ai/DeepSeek-V4-Pro-0813","verifiedAt":"2026-10-06","sources":[{"label":"DeepSeek-V4-Pro-0813","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813"},{"label":"DeepSeek-V4-Pro-0813 LICENSE","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813/blob/main/LICENSE"}]},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/DeepSeek-V4-Pro-0813-GGUF","providerModelId":"unsloth/DeepSeek-V4-Pro-0813-GGUF","verifiedAt":"2026-10-06","sources":[{"label":"DeepSeek-V4-Pro-0813-GGUF","url":"https://huggingface.co/unsloth/DeepSeek-V4-Pro-0813-GGUF"}]},{"kind":"api","label":"Alibaba Cloud Model Studio","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813","providerModelId":"deepseek-v4-pro-0813","verifiedAt":"2026-10-06","sources":[{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]}],"pricing":[{"provider":"Alibaba Cloud Model Studio","providerModelId":"deepseek-v4-pro-0813","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":1.272,"output":3.816,"availability":"active","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"},{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"max","decodeTokensPerSecond":99.1,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"DeepSeek V4 Pro 0813 (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/deepseek-v4-pro/"}]}],"related":[{"id":"deepseek/deepseek-v4.1-flash","name":"DeepSeek-V4.1-Flash","benchmarkUrl":"/benchmarks/models/deepseek-v4.1-flash"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"DeepSeek-V4-Pro-0813","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813"},{"label":"DeepSeek-V4-Pro-0813-GGUF","url":"https://huggingface.co/unsloth/DeepSeek-V4-Pro-0813-GGUF"}]},"deploymentClass":{"value":"rack-scale","status":"inferred","sources":[{"label":"DeepSeek-V4-Pro-0813","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813"},{"label":"DeepSeek-V4-Pro-0813-GGUF","url":"https://huggingface.co/unsloth/DeepSeek-V4-Pro-0813-GGUF"}]},"weightFootprints":[{"artifact":"Unsloth UD-Q4_K_XL GGUF","precisionOrQuantization":"~4-bit","sizeGB":850,"status":"verified","sources":[{"label":"DeepSeek-V4-Pro-0813-GGUF","url":"https://huggingface.co/unsloth/DeepSeek-V4-Pro-0813-GGUF"}]}],"componentStorage":{}},"special":"MIT-licensed reasoning model with a million-token context and 49B active parameters. Its 1.6T-parameter checkpoint targets large GPU servers.","bestUseCase":"Coding agents and research workflows that need long context, with a choice of hosted access or control over the weights.","pricingSummary":{"value":"Alibaba Cloud Model Studio; input 1.272 USD/1M; output 3.816/1M; Global busy-period rate; idle-period rate is lower at $0.636 input / $1.908 output where documented.","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"},{"label":"deepseek-v4-pro-0813 Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-pro-0813"}]},"speedSummary":{"value":"Artificial Analysis: 99.1 output tokens/s (max), observed 2026-10-06.","status":"reviewed","sources":[{"label":"DeepSeek V4 Pro 0813 (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/deepseek-v4-pro/"}]},"relatedArtifacts":[],"normalTokensPerSecond":137.23,"normalTokensPerSecondMin":76.32,"normalTokensPerSecondMax":172.42},"deepseek/deepseek-v4.1-flash":{"id":"deepseek/deepseek-v4.1-flash","name":"DeepSeek-V4.1-Flash","facts":{"family":{"value":"DeepSeek V4.1","status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"}]},"developer":{"value":"DeepSeek-AI","status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"DeepSeek Privacy Policy","url":"https://cdn.deepseek.com/policies/en-US/deepseek-privacy-policy.html"}]},"releasedAt":{"value":"2026-09-10","status":"verified","sources":[{"label":"DeepSeek API change log","url":"https://api-docs.deepseek.com/updates"}]},"license":{"value":"MIT","status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"}],"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/main/LICENSE"},"architecture":{"value":"20-layer encoder + 20-layer decoder with CSA2, Single-Pass mHC, Engram n-gram memory and DSpark/MTP","status":"reviewed","sources":[{"label":"DeepSeek V4.1 Flash model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/README.md"},{"label":"DeepSeek V4.1 technical report","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/resolve/main/DeepSeek_V41_Tech_Report.pdf"},{"label":"DeepSeek checkpoint configuration","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/config.json"},{"label":"Hugging Face checkpoint metadata","url":"https://huggingface.co/api/models/deepseek-ai/DeepSeek-V4.1-Flash"},{"label":"DeepSeek checkpoint index","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/model.safetensors.index.json"},{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"},{"label":"DeepSeek reference inference code","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/inference/generate.py"},{"label":"vLLM Engram implementation","url":"https://docs.vllm.ai/en/latest/api/vllm/models/deepseek_v41/common/engram/"},{"label":"vLLM prefetch implementation","url":"https://github.com/vllm-project/vllm/pull/56512"},{"label":"Experimental NVMe deployment","url":"https://github.com/0xSero/deepseek-v4.1-flash-4x-rtx-pro-6000"},{"label":"NVMe row-store implementation","url":"https://raw.githubusercontent.com/0xSero/deepseek-v4.1-flash-4x-rtx-pro-6000/main/adapter/row_store.cpp"}]},"totalParameters":{"value":763205315794,"status":"reviewed","sources":[{"label":"Hugging Face checkpoint metadata","url":"https://huggingface.co/api/models/deepseek-ai/DeepSeek-V4.1-Flash"},{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"}]},"backboneParameters":{"value":551566195824,"status":"reviewed","sources":[{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"},{"label":"DeepSeek V4.1 Flash model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/README.md"}]},"engramTableParameters":{"value":196613849600,"status":"reviewed","sources":[{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"},{"label":"DeepSeek checkpoint configuration","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/config.json"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"},{"label":"deepseek-v4.1-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"}]},"maxOutputTokens":{"value":393216,"status":"verified","sources":[{"label":"DeepSeek V4.1 Flash model specifications","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"}]},"ngramTableParameters":{"value":196613849600,"status":"reviewed","sources":[{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"},{"label":"DeepSeek checkpoint configuration","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/config.json"}]},"prefillActiveParameters":{"value":8000000000,"status":"reviewed","sources":[{"label":"DeepSeek V4.1 Flash model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/README.md"},{"label":"DeepSeek V4.1 technical report","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/resolve/main/DeepSeek_V41_Tech_Report.pdf"}]},"decodeActiveParameters":{"value":16000000000,"status":"reviewed","sources":[{"label":"DeepSeek V4.1 Flash model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/README.md"},{"label":"DeepSeek V4.1 technical report","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/resolve/main/DeepSeek_V41_Tech_Report.pdf"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/deepseek/deepseek-v4.1-flash","providerModelId":"deepseek/deepseek-v4.1-flash"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"deepseek/deepseek-v4.1-flash"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","providerModelId":"deepseek-ai/DeepSeek-V4.1-Flash","verifiedAt":"2026-10-06","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"}]},{"kind":"api","label":"Alibaba Cloud Model Studio","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash","providerModelId":"deepseek-v4.1-flash","verifiedAt":"2026-10-06","sources":[{"label":"deepseek-v4.1-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"deepseek/deepseek-v4.1-flash","currency":"USD","unit":"per-1M-tokens","input":0.3,"output":1.2,"cachedRead":0.006,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Alibaba Cloud Model Studio","providerModelId":"deepseek-v4.1-flash","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.283,"output":1.131,"cachedRead":0.03,"availability":"active","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"},{"label":"deepseek-v4.1-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"max","decodeTokensPerSecond":226.6,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"DeepSeek V4.1 Flash (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/deepseek-v4-1-flash"}]}],"related":[{"id":"deepseek/deepseek-v4-pro-0813","name":"DeepSeek-V4-Pro-0813","benchmarkUrl":"/benchmarks/models/deepseek-v4-pro-0813"},{"id":"deepseek/deepseek-v4.1-flash-1017","name":"DeepSeek-V4.1-Flash-1017","benchmarkUrl":"/benchmarks/models/deepseek-v4.1-flash-1017"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"},{"label":"deepseek-v4.1-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"},{"label":"DeepSeek V4.1 Flash (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/deepseek-v4-1-flash"}]},"deploymentClass":{"value":"large multi-GPU / rack-scale","status":"inferred","sources":[{"label":"DeepSeek-V4.1-Flash","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash"},{"label":"deepseek-v4.1-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"},{"label":"DeepSeek V4.1 Flash (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/deepseek-v4-1-flash"}]},"weightFootprints":[{"artifact":"Exact released checkpoint tensor payload","precisionOrQuantization":"native mixed released precision","sizeGB":510.286023,"status":"artifact-derived","sources":[{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"}]}],"componentStorage":{"engramNativePayloadGB":{"value":202.7580324,"status":"reviewed","sources":[{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"}]},"remainingNativePayloadGB":{"value":307.5279906,"status":"reviewed","sources":[{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"}]}},"contextMemoryCaveat":{"value":"The released Engram tables occupy about 203GB including scales. Other checkpoint tensors occupy about 308GB, before KV cache and runtime overhead. These are storage footprints, not minimum VRAM requirements.","status":"reviewed","sources":[{"label":"DeepSeek weight files","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/tree/2cba9e42aa026125f3ed06c6d98c1db82f7ca027"},{"label":"DeepSeek V4.1 technical report","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/resolve/main/DeepSeek_V41_Tech_Report.pdf"}]},"officialHostMemoryOffload":{"value":"Engram stores token-indexed vectors that can be prefetched from system RAM.","status":"reviewed","sources":[{"label":"DeepSeek V4.1 technical report","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/resolve/main/DeepSeek_V41_Tech_Report.pdf"},{"label":"vLLM Engram implementation","url":"https://docs.vllm.ai/en/latest/api/vllm/models/deepseek_v41/common/engram/"},{"label":"vLLM prefetch implementation","url":"https://github.com/vllm-project/vllm/pull/56512"}]},"experimentalNvmeOffload":{"value":"An independent NVMe implementation also exists; its reported performance has not been reproduced by us.","status":"reviewed","sources":[{"label":"Experimental NVMe deployment","url":"https://github.com/0xSero/deepseek-v4.1-flash-4x-rtx-pro-6000"},{"label":"NVMe row-store implementation","url":"https://raw.githubusercontent.com/0xSero/deepseek-v4.1-flash-4x-rtx-pro-6000/main/adapter/row_store.cpp"}]},"ngramMemoryPlacement":{"value":"Host RAM offload with asynchronous prefetch is supported; GPU residency depends on the serving engine and configuration.","status":"reviewed","sources":[{"label":"DeepSeek V4.1 technical report","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/resolve/main/DeepSeek_V41_Tech_Report.pdf"}]}},"special":"Pairs a sparse reasoning backbone with 196B parameters of token-indexed Engram memory. Its encoder-decoder design sharply reduces the KV cache needed for long context.","bestUseCase":"Agents that repeatedly read large documents, codebases or conversation histories. Engram host-memory offload is useful when GPU memory is the constraint.","pricingSummary":{"value":"Alibaba Cloud Model Studio; input 0.283 USD/1M; cached read 0.03/1M; output 1.131/1M; Representative busy-period rate; documented idle-period rates are lower and region-specific.","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"},{"label":"deepseek-v4.1-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"}]},"speedSummary":{"value":"Artificial Analysis: 226.6 output tokens/s (max), observed 2026-10-06.","status":"reviewed","sources":[{"label":"DeepSeek V4.1 Flash (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/deepseek-v4-1-flash"}]},"relatedArtifacts":[],"normalTokensPerSecond":241.83,"normalTokensPerSecondMin":195.52,"normalTokensPerSecondMax":311.35},"openai/gpt-6.1-sol":{"id":"openai/gpt-6.1-sol","name":"GPT-6.1 Sol","facts":{"family":{"value":"GPT-6","status":"verified","sources":[{"label":"Introducing GPT-6.1 Sol","url":"https://openai.com/index/introducing-gpt-6-1-sol/"},{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"Introducing GPT-6.1 Sol","url":"https://openai.com/index/introducing-gpt-6-1-sol/"},{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-09-29","status":"verified","sources":[{"label":"Introducing GPT-6.1 Sol","url":"https://openai.com/index/introducing-gpt-6-1-sol/"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-6.1-sol","providerModelId":"openai/gpt-6.1-sol"},{"kind":"api","label":"Official OpenAI model documentation","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol","providerModelId":"gpt-6.1-sol","verifiedAt":"2026-10-06","sources":[{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-6.1-sol","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.1,"cachedWrite":2.5,"availability":"active","status":"verified","sources":[{"label":"Introducing GPT-6.1 Sol","url":"https://openai.com/index/introducing-gpt-6-1-sol/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"max","decodeTokensPerSecond":58,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"GPT-6.1 Sol: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-1-sol"}]}],"related":[{"id":"openai/gpt-6-astra","name":"GPT-6 Astra","benchmarkUrl":"/benchmarks/models/gpt-6-astra"},{"id":"openai/gpt-6-luna","name":"GPT-6 Luna","benchmarkUrl":"/benchmarks/models/gpt-6-luna"},{"id":"openai/gpt-6-sol","name":"GPT-6 Sol","benchmarkUrl":"/benchmarks/models/gpt-6-sol"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"GPT-6.1 Sol: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-1-sol"},{"label":"Introducing GPT-6.1 Sol","url":"https://openai.com/index/introducing-gpt-6-1-sol/"},{"label":"GPT-6.1 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6.1-sol"}]},"weightFootprints":[],"componentStorage":{}},"special":"High-end reasoning and coding with a million-token-class context, at lower token rates than GPT-6 Astra.","bestUseCase":"Difficult coding, analysis and agent tasks where capability matters and Astra's token cost is hard to justify.","pricingSummary":{"value":"OpenAI API; input 2 USD/1M; cached read 0.1/1M; output 10/1M","status":"verified","sources":[{"label":"Introducing GPT-6.1 Sol","url":"https://openai.com/index/introducing-gpt-6-1-sol/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]},"speedSummary":{"value":"Artificial Analysis: 58 output tokens/s (max), observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-6.1 Sol: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-1-sol"}]},"relatedArtifacts":[],"normalTokensPerSecond":51.17,"normalTokensPerSecondMin":48.42,"normalTokensPerSecondMax":54.77},"qwen/qwen3.8-flash":{"id":"qwen/qwen3.8-flash","name":"Qwen3.8-Flash","facts":{"family":{"value":"Qwen 3.8","status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"}]},"developer":{"value":"Qwen / Alibaba Cloud","status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"}]},"origin":{"value":"China (Alibaba Group/Qwen organization context)","status":"verified","sources":[{"label":"Alibaba Group corporate information FAQ","url":"https://www.alibabagroup.com/en-US/faqs-corporate-information"}]},"releasedAt":{"value":"2026-08-26","status":"verified","sources":[{"label":"Model Studio model lifecycle and updates","url":"https://www.alibabacloud.com/help/en/model-studio/newly-released-models"},{"label":"Qwen3.8-Flash-Next repository news","url":"https://github.com/QwenLM/Qwen3.8-Flash-Next"}]},"license":{"value":"Qwen Community 1.0","status":"reviewed","sources":[{"label":"Official Qwen3.8 Flash model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next"},{"label":"Qwen Community License 1.0","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/LICENSE"}],"url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/blob/main/LICENSE"},"architecture":{"value":"Hosted Flash based on Flash-Next: sparse attention and n-gram memory; Flash-Next previews Qwen 4 architecture","status":"reviewed","sources":[{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"},{"label":"Qwen Flash-Next configuration","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/config.json"},{"label":"Qwen Flash-Next release","url":"https://github.com/QwenLM/Qwen3.8-Flash-Next"},{"label":"Qwen 3.8 Flash API","url":"https://www.qwencloud.com/models/qwen3.8-flash"}]},"totalParameters":{"value":180000000000,"status":"inferred","sources":[{"label":"Qwen3.8-Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next"},{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"}]},"maxOutputTokens":{"value":131072,"status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"}]},"weightsAvailability":{"value":"official-open-variants","status":"reviewed","sources":[{"label":"Official Qwen3.8 Flash model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next"},{"label":"Qwen Community License 1.0","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/LICENSE"}]},"activeParameters":{"value":6000000000,"status":"inferred","sources":[{"label":"Qwen3.8-Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/qwen/qwen3.8-flash","providerModelId":"qwen/qwen3.8-flash"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"qwen/qwen3.8-flash"},{"kind":"api","label":"Alibaba Cloud Model Studio","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash","providerModelId":"qwen3.8-flash","verifiedAt":"2026-10-06","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"}]},{"kind":"download","label":"Official Qwen3.8 Flash weights","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next","providerModelId":"Qwen/Qwen3.8-Flash-Next","verifiedAt":"2026-10-08","sources":[{"label":"Official Qwen3.8 Flash model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next"},{"label":"Qwen Community License 1.0","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/LICENSE"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"qwen/qwen3.8-flash","currency":"USD","unit":"per-1M-tokens","input":0.15,"output":0.47,"cachedRead":0.016,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Alibaba Cloud Model Studio","providerModelId":"qwen3.8-flash","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.113,"output":0.382,"cachedRead":0.014,"cachedWrite":0.177,"availability":"active","status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"},{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]}],"speed":[],"related":[{"id":"qwen/qwen3.8-max-0902","name":"Qwen3.8-Max-0902","benchmarkUrl":"/benchmarks/models/qwen3.8-max-0902"},{"id":"qwen/qwen3.8-27B","name":"Qwen3.8-27B","benchmarkUrl":"/benchmarks/models/qwen3.8-27b"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"},{"label":"Qwen3.8-Flash-Next - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-8-flash-next"}]},"weightFootprints":[],"componentStorage":{}},"special":"Hosted Qwen built on Flash-Next, with sparse attention, n-gram memory and a million-token context. Flash-Next previews the architecture planned for Qwen 4.","bestUseCase":"High-volume document, image and video analysis, or coding agents that need long context at low token rates.","pricingSummary":{"value":"Alibaba Cloud Model Studio; input 0.113 USD/1M; cached read 0.014/1M; output 0.382/1M; Representative documented rate; regional prices can differ.","status":"verified","sources":[{"label":"qwen3.8-flash Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-flash"},{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"relatedArtifacts":[{"name":"Qwen3.8-Flash-Next","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next","sources":[{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"},{"label":"Qwen Flash-Next release","url":"https://github.com/QwenLM/Qwen3.8-Flash-Next"}],"facts":{"backboneParameters":{"value":125000000000,"status":"reviewed","sources":[{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"}]},"ngramTableParameters":{"value":51000000000,"status":"reviewed","sources":[{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"}]},"mtpParameters":{"value":4000000000,"status":"reviewed","sources":[{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"}]},"activeParameters":{"value":6000000000,"status":"reviewed","sources":[{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"}]}},"ngramMemoryPlacement":{"value":"Host RAM offload with asynchronous prefetch is supported; GPU residency depends on the serving engine and configuration.","status":"reviewed","sources":[{"label":"Qwen Flash-Next release","url":"https://github.com/QwenLM/Qwen3.8-Flash-Next"}]}}],"relatedArtifactSummary":{"value":"Flash-Next supplies the open weights for the Flash model family and previews the architecture planned for Qwen 4.","status":"reviewed","sources":[{"label":"Qwen Flash-Next model card","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/README.md"},{"label":"Qwen Flash-Next configuration","url":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next/raw/main/config.json"},{"label":"Qwen Flash-Next release","url":"https://github.com/QwenLM/Qwen3.8-Flash-Next"}]},"normalTokensPerSecond":75.13,"normalTokensPerSecondMin":74.24,"normalTokensPerSecondMax":76.81},"moonshotai/kimi-k3":{"id":"moonshotai/kimi-k3","name":"Kimi K3","facts":{"family":{"value":"Kimi K3","status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"},{"label":"Kimi K3 License","url":"https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE"}]},"developer":{"value":"Moonshot AI","status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"},{"label":"Kimi K3 License","url":"https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"About Moonshot AI","url":"https://www.moonshot.cn/about"}]},"releasedAt":{"value":"2026-07-16","status":"reviewed","sources":[{"label":"Kimi K3 tech blog","url":"https://www.kimi.com/blog/kimi-k3"},{"label":"Moonshot launches Kimi K3 with 2.8 trillion parameters and 1M context","url":"https://www.implicator.ai/moonshot-launches-kimi-k3-with-2-8-trillion-parameters-and-1m-context/"}]},"license":{"value":"Kimi K3 License","status":"reviewed","sources":[{"label":"Kimi K3 License","url":"https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE"}],"url":"https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE"},"architecture":{"value":"Mixture-of-Experts with KDA + Gated MLA attention, 896 experts and native vision encoder","status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},"totalParameters":{"value":2800000000000,"status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},"maxOutputTokens":{"value":1048576,"status":"verified","sources":[{"label":"Kimi API chat completion reference","url":"https://platform.kimi.ai/docs/api/chat"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Kimi K3 License","url":"https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE"}]},"activeParameters":{"value":104000000000,"status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/moonshotai/kimi-k3","providerModelId":"moonshotai/kimi-k3"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"moonshotai/kimi-k3"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/moonshotai/Kimi-K3","providerModelId":"moonshotai/Kimi-K3","verifiedAt":"2026-10-06","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/Kimi-K3-GGUF","providerModelId":"unsloth/Kimi-K3-GGUF","verifiedAt":"2026-10-06","sources":[{"label":"Kimi-K3-GGUF","url":"https://huggingface.co/unsloth/Kimi-K3-GGUF"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"moonshotai/kimi-k3","currency":"USD","unit":"per-1M-tokens","input":0.23900000000000002,"output":10.8,"cachedRead":0.21,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"max","decodeTokensPerSecond":44.8,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"Kimi K3 (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/kimi-k3/"}]}],"related":[{"id":"moonshotai/kimi-k3.1-pro","name":"Kimi K3.1 Pro","benchmarkUrl":"/benchmarks/models/kimi-k3.1-pro"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"},{"label":"Kimi-K3-GGUF","url":"https://huggingface.co/unsloth/Kimi-K3-GGUF"}]},"deploymentClass":{"value":"rack-scale","status":"inferred","sources":[{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"},{"label":"Kimi-K3-GGUF","url":"https://huggingface.co/unsloth/Kimi-K3-GGUF"}]},"weightFootprints":[{"artifact":"Unsloth UD-IQ1_M GGUF","precisionOrQuantization":"~1-bit","sizeGB":649,"status":"verified","sources":[{"label":"Kimi-K3-GGUF","url":"https://huggingface.co/unsloth/Kimi-K3-GGUF"}]},{"artifact":"Unsloth UD-Q4_K_XL GGUF","precisionOrQuantization":"~4-bit","sizeGB":1510,"status":"verified","sources":[{"label":"Kimi-K3-GGUF","url":"https://huggingface.co/unsloth/Kimi-K3-GGUF"}]}],"componentStorage":{}},"special":"Open-weight multimodal reasoning at a 2.8T-parameter scale, with a million-token context and 104B parameters active per token.","bestUseCase":"Complex research and agent workflows combining long documents and images. Hosted access avoids the large memory footprint of self-hosting.","speedSummary":{"value":"Artificial Analysis: 44.8 output tokens/s (max), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Kimi K3 (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/kimi-k3/"}]},"relatedArtifacts":[],"normalTokensPerSecond":115.15,"normalTokensPerSecondMin":100.09,"normalTokensPerSecondMax":124.71},"z-ai/glm-5.3":{"id":"z-ai/glm-5.3","name":"GLM-5.3","facts":{"family":{"value":"GLM-5","status":"verified","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"},{"label":"GLM-5.3 License","url":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE"}]},"developer":{"value":"Z.AI","status":"verified","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"},{"label":"GLM-5.3 License","url":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"Zhipu AI company site","url":"https://www.zhipuai.cn/en"},{"label":"Z.ai launches GLM-5.3 with claimed 50% gain on coding benchmark","url":"https://technode.com/2026/08/17/z-ai-launches-glm-5-3-with-claimed-50-gain-on-coding-benchmark/"}]},"releasedAt":{"value":"2026-08-17","status":"reviewed","sources":[{"label":"Z.ai launches GLM-5.3 with claimed 50% gain on coding benchmark","url":"https://technode.com/2026/08/17/z-ai-launches-glm-5-3-with-claimed-50-gain-on-coding-benchmark/"},{"label":"Model Studio model lifecycle and updates","url":"https://www.alibabacloud.com/help/en/model-studio/newly-released-models"},{"label":"Z.AI release notes","url":"https://docs.z.ai/release-notes/new-released"}]},"license":{"value":"GLM-5.3 License","status":"reviewed","sources":[{"label":"GLM-5.3 License","url":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE"}],"url":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE"},"architecture":{"value":"GLM MoE with DSA sparse attention","status":"verified","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"}]},"totalParameters":{"value":753000000000,"status":"verified","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GLM-5.3 developer guide","url":"https://docs.z.ai/guides/llm/glm-5.3"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"GLM-5.3 License","url":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/z-ai/glm-5.3","providerModelId":"z-ai/glm-5.3"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"z-ai/glm-5.3"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/zai-org/GLM-5.3","providerModelId":"zai-org/GLM-5.3","verifiedAt":"2026-10-06","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"},{"label":"GLM-5.3 License","url":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE"}]},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-GGUF","providerModelId":"unsloth/GLM-5.3-GGUF","verifiedAt":"2026-10-06","sources":[{"label":"GLM-5.3-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-GGUF"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash","providerModelId":"zai/glm-5.3","verifiedAt":"2026-10-06","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"z-ai/glm-5.3","currency":"USD","unit":"per-1M-tokens","input":0.079,"output":1.4,"cachedRead":0.078,"fetchedAt":"2026-10-24T12:00:00.000Z","observedAt":"2026-10-24T12:00:00.000Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Z.AI API","providerModelId":"glm-5.3","observedAt":"2026-10-24","currency":"USD","unit":"per-1M-tokens","input":1.4,"output":4.4,"cachedRead":0.26,"availability":"active","status":"verified","sources":[{"label":"Pricing","url":"https://docs.z.ai/guides/overview/pricing"}]},{"provider":"Vercel AI Gateway / provider routes","providerModelId":"zai/glm-5.3","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.7,"output":2.2,"cachedRead":0.12,"availability":"active","status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]}],"speed":[{"provider":"Vercel AI Gateway","observedAt":"2026-10-06","modelSnapshot":"zai/glm-5.3","decodeTokensPerSecond":412,"timeToFirstTokenMs":200,"status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"max","decodeTokensPerSecond":73.3,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"GLM-5.3 (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/glm-5-3/"}]}],"related":[{"id":"ox-alpha","name":"GLM-5.3-Flash","guideUrl":"https://stealthmodels.com/ox-alpha","benchmarkUrl":"/benchmarks/models/glm-5.3-flash"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"},{"label":"GLM-5.3-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-GGUF"},{"label":"GLM-5.3 (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/glm-5-3/"}]},"deploymentClass":{"value":"rack-scale","status":"inferred","sources":[{"label":"GLM-5.3","url":"https://huggingface.co/zai-org/GLM-5.3"},{"label":"GLM-5.3-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-GGUF"},{"label":"GLM-5.3 (max) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/glm-5-3/"}]},"weightFootprints":[{"artifact":"Unsloth UD-Q4_K_XL GGUF","precisionOrQuantization":"~4-bit","sizeGB":467,"status":"verified","sources":[{"label":"GLM-5.3-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-GGUF"}]}],"componentStorage":{}},"special":"Large open-weight reasoning model combining sparse attention with a million-token context. The weights use Z.ai's custom GLM-5.3 license.","bestUseCase":"Text reasoning and coding agents that need long context and deployment control on substantial GPU infrastructure.","pricingSummary":{"value":"Z.AI API; input 1.4 USD/1M; cached read 0.26/1M; output 4.4/1M; cached input storage free for a limited time. Vercel AI Gateway / provider routes; input 0.7 USD/1M; cached read 0.12/1M; output 2.2/1M; Vercel gateway shows multiple provider routes; displayed minimum/current route pricing is not universal Z.AI direct pricing.","status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"speedSummary":{"value":"Vercel AI Gateway: 412 output tokens/s, observed 2026-10-06.","status":"reviewed","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"relatedArtifacts":[],"normalTokensPerSecond":85.97,"normalTokensPerSecondMin":81.86,"normalTokensPerSecondMax":89.16},"openai/gpt-6-astra":{"id":"openai/gpt-6-astra","name":"GPT-6 Astra","facts":{"family":{"value":"GPT-6","status":"verified","sources":[{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"},{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"},{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-09-03","status":"verified","sources":[{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-6-astra","providerModelId":"openai/gpt-6-astra"},{"kind":"api","label":"Official OpenAI model documentation","url":"https://developers.openai.com/api/docs/models/gpt-6-astra","providerModelId":"gpt-6-astra","verifiedAt":"2026-10-06","sources":[{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"},{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-6-astra","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":10,"output":50,"cachedRead":1,"cachedWrite":12.5,"availability":"active","status":"verified","sources":[{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]}],"speed":[{"provider":"OpenAI API","observedAt":"2026-10-06","status":"verified","sources":[{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"release comparison","decodeTokensPerSecond":63.3,"reasoningModeOrBudget":"release comparison","status":"verified","sources":[{"label":"GPT-6 Astra: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-astra"}]}],"related":[{"id":"openai/gpt-6.1-sol","name":"GPT-6.1 Sol","benchmarkUrl":"/benchmarks/models/gpt-6.1-sol"},{"id":"openai/gpt-6-luna","name":"GPT-6 Luna","benchmarkUrl":"/benchmarks/models/gpt-6-luna"},{"id":"openai/gpt-6-sol","name":"GPT-6 Sol","benchmarkUrl":"/benchmarks/models/gpt-6-sol"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"GPT-6 Astra: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-astra"},{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"},{"label":"GPT-6 Astra Model","url":"https://developers.openai.com/api/docs/models/gpt-6-astra"}]},"weightFootprints":[],"componentStorage":{}},"special":"OpenAI's premium GPT-6 tier combines advanced reasoning and computer use with a million-token-class context.","bestUseCase":"Difficult professional and agent tasks where a stronger result is worth a higher token bill.","pricingSummary":{"value":"OpenAI API; input 10 USD/1M; cached read 1/1M; output 50/1M","status":"verified","sources":[{"label":"GPT-6 Astra: A new generation of intelligence","url":"https://openai.com/index/gpt-6-astra/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]},"speedSummary":{"value":"Artificial Analysis: 63.3 output tokens/s (release comparison), observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-6 Astra: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-astra"}]},"relatedArtifacts":[],"normalTokensPerSecond":65.14,"normalTokensPerSecondMin":56.31,"normalTokensPerSecondMax":79.34},"anthropic/claude-opus-5.5":{"id":"anthropic/claude-opus-5.5","name":"Claude Opus 5.5","facts":{"family":{"value":"Claude 5.5","status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"developer":{"value":"Anthropic","status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Anthropic Transparency Hub","url":"https://www.anthropic.com/transparency/voluntary-commitments"}]},"releasedAt":{"value":"2026-09-22","status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Anthropic Commercial Terms of Service","url":"https://www.anthropic.com/legal/commercial-terms"}],"url":"https://www.anthropic.com/legal/commercial-terms"},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/anthropic/claude-opus-5.5","providerModelId":"anthropic/claude-opus-5.5"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"anthropic/claude-opus-5.5"},{"kind":"api","label":"Anthropic API","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview","providerModelId":"claude-opus-5-5","verifiedAt":"2026-10-06","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"anthropic/claude-opus-5.5","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.19999999999999998,"cachedWrite":5,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Anthropic API","providerModelId":"claude-opus-5-5","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.2,"cachedWrite":5,"batchInput":2,"batchOutput":10,"availability":"active","status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]}],"speed":[{"provider":"Anthropic","observedAt":"2026-10-06","status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"release comparison","decodeTokensPerSecond":94,"reasoningModeOrBudget":"release comparison","status":"verified","sources":[{"label":"Claude Opus 5.5: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/claude-opus-5-5"}]}],"related":[{"id":"anthropic/claude-sonnet-5.5","name":"Claude Sonnet 5.5","benchmarkUrl":"/benchmarks/models/claude-sonnet-5.5"},{"id":"anthropic/claude-haiku-5.5","name":"Claude Haiku 5.5","benchmarkUrl":"/benchmarks/models/claude-haiku-5.5"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Claude Opus 5.5: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/claude-opus-5-5"},{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"weightFootprints":[],"componentStorage":{}},"special":"Anthropic's top Opus tier focuses on complex reasoning and coding, with a million-token context for large tasks.","bestUseCase":"Long-running coding agents and demanding professional analysis where resolving the task matters more than minimizing token cost.","pricingSummary":{"value":"Anthropic API; input 4 USD/1M; cached read 0.2/1M; output 20/1M; 5-minute cache write $5/M; 1-hour cache write $8/M. Batch rates shown as 50% of standard.","status":"verified","sources":[{"label":"Claude Opus 5.5","url":"https://platform.claude.com/docs/en/models/opus-5-5/overview"}]},"speedSummary":{"value":"Artificial Analysis: 94 output tokens/s (release comparison), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Claude Opus 5.5: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/claude-opus-5-5"}]},"relatedArtifacts":[],"normalTokensPerSecond":165.84,"normalTokensPerSecondMin":159.4,"normalTokensPerSecondMax":178.09},"anthropic/claude-fable-5.1":{"id":"anthropic/claude-fable-5.1","name":"Claude Fable 5.1","facts":{"family":{"value":"Claude 5.1","status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"developer":{"value":"Anthropic","status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Anthropic Transparency Hub","url":"https://www.anthropic.com/transparency/voluntary-commitments"}]},"releasedAt":{"value":"2026-09-01","status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Anthropic Commercial Terms of Service","url":"https://www.anthropic.com/legal/commercial-terms"}],"url":"https://www.anthropic.com/legal/commercial-terms"},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/anthropic/claude-fable-5.1","providerModelId":"anthropic/claude-fable-5.1"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"anthropic/claude-fable-5.1"},{"kind":"api","label":"Anthropic API","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview","providerModelId":"claude-fable-5-1","verifiedAt":"2026-10-06","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"anthropic/claude-fable-5.1","currency":"USD","unit":"per-1M-tokens","input":10,"output":50,"cachedRead":0.25,"cachedWrite":12.5,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Anthropic API","providerModelId":"claude-fable-5-1","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":10,"output":50,"cachedRead":0.25,"cachedWrite":12.5,"batchInput":5,"batchOutput":25,"availability":"active","status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]}],"speed":[{"provider":"Anthropic","observedAt":"2026-10-06","status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":61.4,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Claude Fable 5.1 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/claude-fable-5-1/"}]}],"related":[],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Claude Fable 5.1 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/claude-fable-5-1/"},{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"weightFootprints":[],"componentStorage":{}},"special":"A premium Claude reasoning model with a million-token context. Its measured output is slower and more verbose than Sonnet 5.5.","bestUseCase":"Difficult reasoning and agent work where extended deliberation is useful and response latency is acceptable.","pricingSummary":{"value":"Anthropic API; input 10 USD/1M; cached read 0.25/1M; output 50/1M; 1-hour cache write $20/M; batch 50% discount.","status":"verified","sources":[{"label":"Claude Fable 5.1","url":"https://platform.claude.com/docs/en/models/fable-5-1/overview"}]},"speedSummary":{"value":"Artificial Analysis: 61.4 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Claude Fable 5.1 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/claude-fable-5-1/"}]},"relatedArtifacts":[],"normalTokensPerSecond":122.12,"normalTokensPerSecondMin":121.83,"normalTokensPerSecondMax":122.43},"meta/muse-spark-1.3":{"id":"meta/muse-spark-1.3","name":"Muse Spark 1.3","facts":{"family":{"value":"Muse Spark","status":"verified","sources":[{"label":"Introducing Muse Spark 1.3","url":"https://research.meta.ai/blog/introducing-muse-spark-1-3"},{"label":"Models - Meta Model API","url":"https://dev.meta.ai/docs/models"}]},"developer":{"value":"Meta","status":"verified","sources":[{"label":"Introducing Muse Spark 1.3","url":"https://research.meta.ai/blog/introducing-muse-spark-1-3"},{"label":"Models - Meta Model API","url":"https://dev.meta.ai/docs/models"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Moving to a Local Selling Model","url":"https://about.fb.com/news/2017/12/moving-to-a-local-selling-model/"}]},"releasedAt":{"value":"2026-09-02","status":"verified","sources":[{"label":"Introducing Muse Spark 1.3","url":"https://research.meta.ai/blog/introducing-muse-spark-1-3"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Meta Model API Terms of Service","url":"https://dev.meta.ai/legal/terms-of-service"}],"url":"https://dev.meta.ai/legal/terms-of-service"},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Models - Meta Model API","url":"https://dev.meta.ai/docs/models"}]},"inputModalities":{"value":["text","image","video","audio","PDF"],"status":"verified","sources":[{"label":"Models - Meta Model API","url":"https://dev.meta.ai/docs/models"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Models - Meta Model API","url":"https://dev.meta.ai/docs/models"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/meta/muse-spark-1.3","providerModelId":"meta/muse-spark-1.3"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"meta/muse-spark-1.3"},{"kind":"api","label":"Meta Model API","url":"https://dev.meta.ai/docs/models","providerModelId":"muse-spark-1.3","verifiedAt":"2026-10-06","sources":[{"label":"Models - Meta Model API","url":"https://dev.meta.ai/docs/models"},{"label":"Meta Model API","url":"https://dev.meta.ai/products/meta-model-api"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"meta/muse-spark-1.3","currency":"USD","unit":"per-1M-tokens","input":1.25,"output":4.25,"cachedRead":0.15,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Meta Model API","providerModelId":"muse-spark-1.3","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":1.25,"output":4.25,"cachedRead":0.15,"availability":"active","status":"verified","sources":[{"label":"Meta Model API","url":"https://dev.meta.ai/products/meta-model-api"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":134.5,"timeToFirstTokenMs":53300,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Muse Spark 1.3 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/muse-spark-1-3"}]}],"related":[{"id":"meta/muse-wonder","name":"Muse Wonder","benchmarkUrl":"/benchmarks/models/muse-wonder"},{"id":"meta/muse-spark-open","name":"Muse Spark Open","benchmarkUrl":"/benchmarks/models/muse-spark-open"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Muse Spark 1.3 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/muse-spark-1-3"},{"label":"Introducing Muse Spark 1.3","url":"https://research.meta.ai/blog/introducing-muse-spark-1-3"}]},"weightFootprints":[],"componentStorage":{}},"special":"Meta's hosted reasoning model handles text, images and video within a million-token-class context through an OpenAI-compatible API.","bestUseCase":"Multimodal analysis and coding agents that need Meta's hosted models without running an open-weight checkpoint.","pricingSummary":{"value":"Meta Model API; input 1.25 USD/1M; cached read 0.15/1M; output 4.25/1M; Standard tier; prompts/completions are not used to improve Meta products. Contributor variant is a distinct provider model ID and is not substituted here.","status":"verified","sources":[{"label":"Meta Model API","url":"https://dev.meta.ai/products/meta-model-api"}]},"speedSummary":{"value":"Artificial Analysis: 134.5 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Muse Spark 1.3 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/muse-spark-1-3"}]},"relatedArtifacts":[],"normalTokensPerSecond":321.12,"normalTokensPerSecondMin":262.87,"normalTokensPerSecondMax":398.93},"tencent/hy4-preview":{"id":"tencent/hy4-preview","name":"Hy4 preview","facts":{"family":{"value":"Hy4","status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"developer":{"value":"Tencent Hy Team","status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"About Tencent","url":"https://www.tencent.com/en-us/about.html.html"}]},"releasedAt":{"value":"2026-08-28","status":"reviewed","sources":[{"label":"Tencent open-sources Hy4 preview with 770B parameters and a 1M-token context","url":"https://technode.com/2026/08/28/tencent-open-sources-hy4-preview-with-770b-parameters-and-a-1m-token-context/"},{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"}]},"license":{"value":"Apache-2.0","status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}],"url":"https://github.com/Tencent-Hunyuan/Hy4-preview/blob/main/LICENSE"},"architecture":{"value":"MoE; 78-layer backbone with Gated DSA, 256 routed experts + 1 shared expert, top-8 routed experts active, plus native MTP layer","status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"totalParameters":{"value":770000000000,"status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"maxOutputTokens":{"value":64000,"status":"verified","sources":[{"label":"OpenRouter Hy4 preview endpoints","url":"https://openrouter.ai/api/v1/models/tencent/hy4-preview/endpoints"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},"activeParameters":{"value":49000000000,"status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/tencent/hy4-preview","providerModelId":"tencent/hy4-preview"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"tencent/hy4-preview"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/tencent/Hy4-preview","providerModelId":"tencent/Hy4-preview","verifiedAt":"2026-10-06","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]},{"kind":"download","label":"Official FP8 Hugging Face weights","url":"https://huggingface.co/tencent/Hy4-preview-FP8","providerModelId":"tencent/Hy4-preview-FP8","verifiedAt":"2026-10-06","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"}]}],"pricing":[{"provider":"Tencent","providerModelId":"tencent/hy4-preview","observedAt":"2026-10-23","currency":"USD","unit":"per-1M-tokens","input":0.834,"output":2.501,"cachedRead":0.042,"availability":"active","status":"verified","sources":[{"label":"OpenRouter Hy4 preview endpoints","url":"https://openrouter.ai/api/v1/models/tencent/hy4-preview/endpoints"},{"label":"Tencent open-sources Hy4 preview with 770B parameters and a 1M-token context","url":"https://technode.com/2026/08/28/tencent-open-sources-hy4-preview-with-770b-parameters-and-a-1m-token-context/"}]}],"speed":[],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"},{"label":"Hy4-preview-FP8 weights","url":"https://huggingface.co/tencent/Hy4-preview-FP8/tree/main"}]},"deploymentClass":{"value":"rack-scale","status":"inferred","sources":[{"label":"Tencent-Hunyuan/Hy4-preview","url":"https://github.com/Tencent-Hunyuan/Hy4-preview"},{"label":"Hy4-preview-FP8 weights","url":"https://huggingface.co/tencent/Hy4-preview-FP8/tree/main"}]},"weightFootprints":[{"artifact":"Official Hy4-preview-FP8 repository","precisionOrQuantization":"FP8","sizeGB":814,"status":"verified","sources":[{"label":"Hy4-preview-FP8 weights","url":"https://huggingface.co/tencent/Hy4-preview-FP8/tree/main"}]}],"componentStorage":{}},"special":"Apache-2.0 MoE preview with 770B total parameters, 49B active and a million-token context. Sparse attention and multi-token prediction support long-input workloads. Very heavy in thinking on each instruction turn.","bestUseCase":"Long-document analysis and research on large GPU systems, with the freedom to inspect and adapt the weights.","pricingSummary":{"value":"Tencent API; input 0.834 USD/1M; cached read 0.042/1M; output 2.501/1M; 10% lower from 16:00 to 24:00 UTC","status":"verified","sources":[{"label":"OpenRouter Hy4 preview endpoints","url":"https://openrouter.ai/api/v1/models/tencent/hy4-preview/endpoints"},{"label":"Tencent open-sources Hy4 preview with 770B parameters and a 1M-token context","url":"https://technode.com/2026/08/28/tencent-open-sources-hy4-preview-with-770b-parameters-and-a-1m-token-context/"}]},"relatedArtifacts":[],"normalTokensPerSecond":63.95,"normalTokensPerSecondMin":20.89,"normalTokensPerSecondMax":89.34},"nvidia/nemotron-3-ultra-550b-a55b":{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"Nemotron 3 Ultra","facts":{"family":{"value":"Nemotron 3 Ultra","status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"},{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]},"developer":{"value":"NVIDIA","status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"},{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Contact NVIDIA","url":"https://www.nvidia.com/en-us/contact/"}]},"releasedAt":{"value":"2026-06-04","status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}]},"license":{"value":"OpenMDW License Agreement v1.1","status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}],"url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4"},"architecture":{"value":"LatentMoE: Mamba-2 + MoE + Attention hybrid with Multi-Token Prediction","status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}]},"totalParameters":{"value":550000000000,"status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"},{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"},{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}]},"activeParameters":{"value":55000000000,"status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/nvidia/nemotron-3-ultra-550b-a55b","providerModelId":"nvidia/nemotron-3-ultra-550b-a55b"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"nvidia/nemotron-3-ultra-550b-a55b"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16","providerModelId":"nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16"},{"kind":"api","label":"NVIDIA NIM free endpoint","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b","providerModelId":"nvidia/nemotron-3-ultra-550b-a55b","verifiedAt":"2026-10-06","sources":[{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]},{"kind":"download","label":"Official NVFP4 weights","url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4","providerModelId":"nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4","verifiedAt":"2026-10-06","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"nvidia/nemotron-3-ultra-550b-a55b","currency":"USD","unit":"per-1M-tokens","input":0.6,"output":2.4,"cachedRead":0.12,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"OpenRouter via NVIDIA Build","providerModelId":"nvidia/nemotron-3-ultra-550b-a55b","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.5,"output":2.2,"availability":"active","status":"verified","sources":[{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]},{"provider":"DigitalOcean via NVIDIA Build","providerModelId":"nvidia/nemotron-3-ultra-550b-a55b","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.9,"output":1.7,"availability":"active","status":"verified","sources":[{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":176.9,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Nemotron 3 Ultra 550B A55B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/nvidia-nemotron-3-ultra-550b-a55b/"}]}],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"},{"label":"NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4","url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4"}]},"deploymentClass":{"value":"multi-GPU datacenter","status":"inferred","sources":[{"label":"NVIDIA Nemotron 3 Ultra model card","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b/modelcard"},{"label":"NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4","url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4"}]},"weightFootprints":[],"componentStorage":{}},"special":"Combines Mamba, attention and mixture-of-experts layers in a 550B-parameter model, with 55B active and multi-token prediction for faster serving.","bestUseCase":"Self-hosted reasoning, tool use and retrieval over long documents on multi-GPU NVIDIA infrastructure.","pricingSummary":{"value":"OpenRouter via NVIDIA Build; input 0.5 USD/1M; output 2.2/1M; Partner endpoint price displayed by NVIDIA Build; not a universal model price.","status":"verified","sources":[{"label":"Nemotron 3 Ultra NIM","url":"https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b"}]},"speedSummary":{"value":"Artificial Analysis: 176.9 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Nemotron 3 Ultra 550B A55B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/nvidia-nemotron-3-ultra-550b-a55b/"}]},"relatedArtifacts":[],"normalTokensPerSecond":139.44,"normalTokensPerSecondMin":125.86,"normalTokensPerSecondMax":147.58},"qwen/qwen3.8-max-0902":{"id":"qwen/qwen3.8-max-0902","name":"Qwen3.8-Max-0902","facts":{"family":{"value":"Qwen 3.8","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"developer":{"value":"Qwen / Alibaba Cloud","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"origin":{"value":"China (Alibaba Group/Qwen organization context)","status":"verified","sources":[{"label":"Alibaba Group corporate information FAQ","url":"https://www.alibabagroup.com/en-US/faqs-corporate-information"}]},"releasedAt":{"value":"2026-09-02","status":"verified","sources":[{"label":"Model Studio model lifecycle and updates","url":"https://www.alibabacloud.com/help/en/model-studio/newly-released-models"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Model Studio related agreements","url":"https://www.alibabacloud.com/help/en/model-studio/related-agreements"},{"label":"Model Studio model lifecycle and updates","url":"https://www.alibabacloud.com/help/en/model-studio/newly-released-models"}],"url":"https://www.alibabacloud.com/help/en/model-studio/related-agreements"},"architecture":{"value":"Mixture-of-Experts; hosted version based on the open Qwen3.8-2.4T-A95B","status":"inferred","sources":[{"label":"Model Studio model lifecycle and updates","url":"https://www.alibabacloud.com/help/en/model-studio/newly-released-models"},{"label":"Qwen3.8-2.4T-A95B model card","url":"https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B"}]},"totalParameters":{"value":2400000000000,"status":"inferred","sources":[{"label":"Model Studio model lifecycle and updates","url":"https://www.alibabacloud.com/help/en/model-studio/newly-released-models"},{"label":"Qwen3.8-2.4T-A95B model card","url":"https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"maxOutputTokens":{"value":131072,"status":"inferred","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"activeParameters":{"value":95000000000,"status":"inferred","sources":[{"label":"Qwen3.8-2.4T-A95B model card","url":"https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/qwen/qwen3.8-max-0902","providerModelId":"qwen/qwen3.8-max-0902"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"qwen/qwen3.8-max-0902"},{"kind":"api","label":"Alibaba Cloud Model Studio","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","providerModelId":"qwen3.8-max-0902","verifiedAt":"2026-10-06","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"qwen/qwen3.8-max-0902","currency":"USD","unit":"per-1M-tokens","input":2,"output":6,"cachedRead":0.25,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Alibaba Cloud Model Studio","providerModelId":"qwen3.8-max-0902","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":1.65,"output":4.951,"availability":"active","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"0902","decodeTokensPerSecond":39.3,"timeToFirstTokenMs":2940,"reasoningModeOrBudget":"0902","status":"verified","sources":[{"label":"Qwen3.8 Max (0902) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-8-max"}]}],"related":[{"id":"qwen/qwen3.8-flash","name":"Qwen3.8-Flash","benchmarkUrl":"/benchmarks/models/qwen3.8-flash"},{"id":"qwen/qwen3.8-27B","name":"Qwen3.8-27B","benchmarkUrl":"/benchmarks/models/qwen3.8-27b"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Qwen3.8 Max (0902) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-8-max"}]},"weightFootprints":[],"componentStorage":{}},"special":"A dated Qwen Max snapshot with a million-token-class context, suited to applications that need an explicit model version.","bestUseCase":"Long-context reasoning and agent workloads where a pinned snapshot makes evaluation and deployment easier to reproduce.","pricingSummary":{"value":"Alibaba Cloud Model Studio; input 1.65 USD/1M; output 4.951/1M; Documented 0-1M context tier; regional/tiered prices may differ.","status":"verified","sources":[{"label":"Model Studio model pricing","url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing"}]},"speedSummary":{"value":"Artificial Analysis: 39.3 output tokens/s (0902), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Qwen3.8 Max (0902) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-8-max"}]},"relatedArtifacts":[],"normalTokensPerSecond":75.69,"normalTokensPerSecondMin":63.95,"normalTokensPerSecondMax":90.94},"qwen/qwen3.6-35b-a3b":{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen3.6-35B-A3B","facts":{"family":{"value":"Qwen 3.6","status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"},{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]},"developer":{"value":"Qwen","status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"},{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]},"origin":{"value":"China (Alibaba Group/Qwen organization context)","status":"verified","sources":[{"label":"Alibaba Group corporate information FAQ","url":"https://www.alibabagroup.com/en-US/faqs-corporate-information"}]},"releasedAt":{"value":"2026-04-16","status":"verified","sources":[{"label":"Qwen3.8 repository news","url":"https://github.com/QwenLM/Qwen3.8"}]},"license":{"value":"Apache-2.0","status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"}],"url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"},"architecture":{"value":"MoE causal language model with vision encoder; Gated DeltaNet + gated attention; 256 experts, 8 routed + 1 shared active","status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"}]},"totalParameters":{"value":35000000000,"status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"}]},"contextWindowTokens":{"value":1010000,"status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"}]},"maxOutputTokens":{"value":65536,"status":"verified","sources":[{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"},{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"},{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"}]},"activeParameters":{"value":3000000000,"status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/qwen/qwen3.6-35b-a3b","providerModelId":"qwen/qwen3.6-35b-a3b"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"qwen/qwen3.6-35b-a3b"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B","providerModelId":"Qwen/Qwen3.6-35B-A3B","verifiedAt":"2026-10-06","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"}]},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF","providerModelId":"unsloth/Qwen3.6-35B-A3B-GGUF"},{"kind":"api","label":"Alibaba Cloud Model Studio","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b","providerModelId":"qwen3.6-35b-a3b","verifiedAt":"2026-10-06","sources":[{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"qwen/qwen3.6-35b-a3b","currency":"USD","unit":"per-1M-tokens","input":0.09999999999999999,"output":0.95,"cachedRead":0.049999999999999996,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Alibaba Cloud Model Studio","providerModelId":"qwen3.6-35b-a3b","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.248,"output":1.485,"availability":"active","status":"verified","sources":[{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":135.3,"timeToFirstTokenMs":2110,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Qwen3.6 35B A3B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-6-35b-a3b/"}]}],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"},{"label":"Qwen3.6-35B-A3B-FP8","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B-FP8"}]},"deploymentClass":{"value":"high-memory single GPU / small multi-GPU","status":"inferred","sources":[{"label":"Qwen3.6-35B-A3B","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B"},{"label":"Qwen3.6-35B-A3B-FP8","url":"https://huggingface.co/Qwen/Qwen3.6-35B-A3B-FP8"}]},"weightFootprints":[],"componentStorage":{}},"special":"A multimodal 35B-parameter MoE that activates 3B parameters per token. It combines open Apache-2.0 weights with an emphasis on coding agents.","bestUseCase":"Local or hosted coding assistants that need image understanding and modest active compute. Plan memory for the full checkpoint.","pricingSummary":{"value":"Alibaba Cloud Model Studio; input 0.248 USD/1M; output 1.485/1M; Representative documented rate; region-specific prices exist.","status":"verified","sources":[{"label":"qwen3.6-35b-a3b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-35b-a3b"}]},"speedSummary":{"value":"Artificial Analysis: 135.3 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Qwen3.6 35B A3B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-6-35b-a3b/"}]},"relatedArtifacts":[],"normalTokensPerSecond":577.38,"normalTokensPerSecondMin":576.23,"normalTokensPerSecondMax":578.32},"qwen/qwen3.5-9b":{"id":"qwen/qwen3.5-9b","name":"Qwen3.5-9B","facts":{"family":{"value":"Qwen 3.5","status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"}]},"developer":{"value":"Qwen","status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"}]},"origin":{"value":"China (Alibaba Group/Qwen organization context)","status":"verified","sources":[{"label":"Alibaba Group corporate information FAQ","url":"https://www.alibabagroup.com/en-US/faqs-corporate-information"}]},"releasedAt":{"value":"2026-03-02","status":"verified","sources":[{"label":"Qwen3.8 repository news","url":"https://github.com/QwenLM/Qwen3.8"}]},"license":{"value":"Apache-2.0","status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"}],"url":"https://huggingface.co/Qwen/Qwen3.5-9B"},"architecture":{"value":"Dense causal language model with vision encoder; Gated DeltaNet + gated attention; MTP-trained","status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"},{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},"totalParameters":{"value":9000000000,"status":"verified","sources":[{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},"contextWindowTokens":{"value":1010000,"status":"verified","sources":[{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"},{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"},{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"}]},"activeParameters":{"value":9000000000,"status":"inferred","sources":[{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/qwen/qwen3.5-9b","providerModelId":"qwen/qwen3.5-9b"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"qwen/qwen3.5-9b"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/Qwen/Qwen3.5-9B","providerModelId":"Qwen/Qwen3.5-9B","verifiedAt":"2026-10-06","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"}]},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF","providerModelId":"unsloth/Qwen3.5-9B-GGUF","verifiedAt":"2026-10-06","sources":[{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"qwen/qwen3.5-9b","currency":"USD","unit":"per-1M-tokens","input":0.16999999999999998,"output":0.25,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":65,"timeToFirstTokenMs":1600,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Qwen3.5 9B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-5-9b"}]}],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"},{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},"deploymentClass":{"value":"consumer/workstation","status":"inferred","sources":[{"label":"Qwen3.5-9B","url":"https://huggingface.co/Qwen/Qwen3.5-9B"},{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},"weightFootprints":[{"artifact":"Unsloth Q8_0 GGUF","precisionOrQuantization":"8-bit","sizeGB":9.53,"status":"verified","sources":[{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},{"artifact":"Unsloth BF16 GGUF","precisionOrQuantization":"BF16","sizeGB":17.9,"status":"verified","sources":[{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]},{"artifact":"Multimodal projector","precisionOrQuantization":"BF16","sizeGB":0.922,"status":"verified","sources":[{"label":"Qwen3.5-9B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.5-9B-GGUF"}]}],"componentStorage":{}},"special":"A compact, Apache-2.0 multimodal model with Q4 weight files around 5–6 GB, making local deployment practical on smaller machines.","bestUseCase":"Local assistants, document and image understanding, and lightweight tool workflows on consumer hardware.","speedSummary":{"value":"Artificial Analysis: 65 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Qwen3.5 9B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-5-9b"}]},"relatedArtifacts":[],"normalTokensPerSecond":161.15,"normalTokensPerSecondMin":110.93,"normalTokensPerSecondMax":231.31},"inclusionai/ling-3.1-flash":{"id":"inclusionai/ling-3.1-flash","name":"Ling 3.1 Flash","facts":{"family":{"value":"Ling 3.1","status":"verified","sources":[{"label":"Ling 3.1 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash"}]},"developer":{"value":"InclusionAI","status":"verified","sources":[{"label":"Ling 3.1 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash"}]},"origin":{"value":"China","status":"reviewed","sources":[{"label":"inclusionAI on Hugging Face","url":"https://huggingface.co/inclusionAI"},{"label":"inclusionAI GitHub organization","url":"https://github.com/inclusionAI"},{"label":"Ant Group","url":"https://en.wikipedia.org/wiki/Ant_Group"}]},"releasedAt":{"value":"2026-09-29","status":"verified","sources":[{"label":"Ling 3.1 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash"}]},"license":{"value":"MIT","status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"}],"url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"},"architecture":{"value":"Hybrid-linear MoE; 72 layers (63 KDA, 9 gated MLA); 512 routed experts, 8 selected and 1 shared.","status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"},{"label":"Ling 3.1 Flash released configuration","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash/blob/a4ccd6b45c002ddb236f9cb77d52cc427482bf19/config.json"}]},"totalParameters":{"value":560000000000,"status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"}]},"contextWindowTokens":{"value":262144,"status":"verified","sources":[{"label":"Ling 3.1 Flash released configuration","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash/blob/a4ccd6b45c002ddb236f9cb77d52cc427482bf19/config.json"},{"label":"Ling 3.1 Flash OpenRouter endpoint specifications","url":"https://openrouter.ai/api/v1/models/inclusionai/ling-3.1-flash/endpoints"},{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Ling 3.1 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Ling 3.1 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"},{"label":"Ling 3.1 Flash released configuration","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash/blob/a4ccd6b45c002ddb236f9cb77d52cc427482bf19/config.json"}]},"activeParameters":{"value":25000000000,"status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/inclusionai/ling-3.1-flash","providerModelId":"inclusionai/ling-3.1-flash","verifiedAt":"2026-10-16","sources":[{"label":"Ling 3.1 Flash OpenRouter endpoint specifications","url":"https://openrouter.ai/api/v1/models/inclusionai/ling-3.1-flash/endpoints"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash","providerModelId":"inclusionai/ling-3.1-flash","verifiedAt":"2026-10-16","sources":[{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},{"kind":"download","label":"Official FP8 weights on Hugging Face","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash","providerModelId":"inclusionAI/Ling-3.1-flash","verifiedAt":"2026-10-16","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"}]}],"pricing":[{"provider":"Vercel AI Gateway","providerModelId":"inclusionai/ling-3.1-flash","currency":"USD","unit":"per-1M-tokens","input":0.3,"output":0.8999999999999999,"cachedRead":0.06,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},{"provider":"OpenRouter","providerModelId":"inclusionai/ling-3.1-flash","currency":"USD","unit":"per-1M-tokens","input":0.075,"output":0.22499999999999998,"cachedRead":0.015,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]}],"speed":[{"provider":"Vercel AI Gateway / Novita AI","observedAt":"2026-10-06","modelSnapshot":"inclusionai/ling-3.1-flash","decodeTokensPerSecond":67,"timeToFirstTokenMs":3300,"status":"verified","sources":[{"label":"Ling 3.1 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":210.7,"timeToFirstTokenMs":1800,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Ling 3.1 Flash - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/ling-3-1-flash"}]}],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"},{"label":"Ling 3.1 Flash released configuration","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash/blob/a4ccd6b45c002ddb236f9cb77d52cc427482bf19/config.json"}]},"deploymentClass":{"value":"Multi-GPU server; documented SGLang deployments on 8 H200 or 4 GB300 GPUs.","status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"},{"label":"Ling 3.1 Flash SGLang deployment cookbook","url":"https://docs.sglang.io/cookbook/autoregressive/InclusionAI/Ling-3.1-flash"}]},"weightFootprints":[],"componentStorage":{},"contextMemoryCaveat":{"value":"The developer advertises 1M context; the released configuration and current gateways specify 262,144 tokens. OpenRouter lists up to 235,929 output tokens; Vercel caps output at 32,768. Self-hosting requires memory for the weights, runtime and the selected workload.","status":"verified","sources":[{"label":"Ling 3.1 Flash official model card","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash"},{"label":"Ling 3.1 Flash released configuration","url":"https://huggingface.co/inclusionAI/Ling-3.1-flash/blob/a4ccd6b45c002ddb236f9cb77d52cc427482bf19/config.json"},{"label":"Ling 3.1 Flash OpenRouter endpoint specifications","url":"https://openrouter.ai/api/v1/models/inclusionai/ling-3.1-flash/endpoints"},{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]}},"special":"MIT-licensed hybrid reasoning MoE with 560B total and 25B active parameters, released FP8 weights, coding and tool-use support.","bestUseCase":"Coding, multi-step analysis and tool-using agents through hosted gateways or a high-memory multi-GPU deployment.","pricingSummary":{"value":"Per 1M tokens: OpenRouter / Novita $0.075 input, $0.225 output and $0.015 cached input at current discounted rates; Vercel $0.30 input, $0.90 output and $0.06 cached input. Observed 2026-10-16.","status":"verified","sources":[{"label":"Ling 3.1 Flash OpenRouter endpoint specifications","url":"https://openrouter.ai/api/v1/models/inclusionai/ling-3.1-flash/endpoints"},{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},"speedSummary":{"value":"Vercel AI Gateway / Novita AI: 67 output tokens/s, observed 2026-10-06.","status":"reviewed","sources":[{"label":"Ling 3.1 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/ling-3.1-flash"}]},"relatedArtifacts":[],"normalTokensPerSecond":114.92,"normalTokensPerSecondMin":108.66,"normalTokensPerSecondMax":118.45},"Aleph-Alpha/Kolibri-1":{"id":"Aleph-Alpha/Kolibri-1","name":"Kolibri-1","facts":{"family":{"value":"Kolibri","status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"},{"label":"Kolibri Has Landed: A Sovereign Open-Weight Model","url":"https://aleph-alpha.com/en/blog/kolibri-has-landed-a-sovereign-open-weight-model/"}]},"developer":{"value":"Aleph Alpha","status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"},{"label":"Kolibri Has Landed: A Sovereign Open-Weight Model","url":"https://aleph-alpha.com/en/blog/kolibri-has-landed-a-sovereign-open-weight-model/"}]},"origin":{"value":"Germany","status":"verified","sources":[{"label":"Reto Spörri leaves Aleph Alpha, Ilhan Scheer continues as CEO","url":"https://aleph-alpha.com/en/news/reto-spoerri-leaves-ilhan-scheer-ceo/"}]},"releasedAt":{"value":"2026-10-03","status":"verified","sources":[{"label":"Kolibri Has Landed: A Sovereign Open-Weight Model","url":"https://aleph-alpha.com/en/blog/kolibri-has-landed-a-sovereign-open-weight-model/"}]},"license":{"value":"Apache-2.0","status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"},{"label":"Kolibri Has Landed: A Sovereign Open-Weight Model","url":"https://aleph-alpha.com/en/blog/kolibri-has-landed-a-sovereign-open-weight-model/"}],"url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"},"architecture":{"value":"Mixture-of-Experts with 4:1 sliding-window-attention to global-attention pattern","status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]},"totalParameters":{"value":78103074560,"status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]},"contextWindowTokens":{"value":1048576,"status":"reviewed","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"},{"label":"Kolibri Has Landed: A Sovereign Open-Weight Model","url":"https://aleph-alpha.com/en/blog/kolibri-has-landed-a-sovereign-open-weight-model/"}]},"activeParameters":{"value":3457573120,"status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]}},"access":[{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1","providerModelId":"Aleph-Alpha/Kolibri-1","verifiedAt":"2026-10-06","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]}],"pricing":[],"speed":[],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]},"deploymentClass":{"value":"single/dual datacenter GPU","status":"inferred","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]},"weightFootprints":[{"artifact":"Official model memory footprint","precisionOrQuantization":"FP8 mixed with selected BF16 tensors","sizeGB":78,"status":"verified","sources":[{"label":"Kolibri-1","url":"https://huggingface.co/Aleph-Alpha/Kolibri-1"}]}],"componentStorage":{}},"special":"German-English MoE with Apache-2.0 weights and about 3.46B active parameters. Trained at 262k context, it has been validated out to one million tokens.","bestUseCase":"Self-hosted German and English document analysis, retrieval and tool workflows on a high-memory GPU server.","relatedArtifacts":[],"normalTokensPerSecond":165.76,"normalTokensPerSecondMin":145.48,"normalTokensPerSecondMax":184.57},"anthropic/claude-haiku-5.5":{"id":"anthropic/claude-haiku-5.5","name":"Claude Haiku 5.5","facts":{"family":{"value":"Claude 5.5","status":"verified","sources":[{"label":"Introducing Claude Haiku 5.5","url":"https://www.anthropic.com/claude-haiku-5-5"}]},"developer":{"value":"Anthropic","status":"verified","sources":[{"label":"Introducing Claude Haiku 5.5","url":"https://www.anthropic.com/claude-haiku-5-5"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Anthropic Transparency Hub","url":"https://www.anthropic.com/transparency/voluntary-commitments"}]},"releasedAt":{"value":"2026-10-07","status":"verified","sources":[{"label":"Introducing Claude Haiku 5.5","url":"https://www.anthropic.com/claude-haiku-5-5"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Anthropic Commercial Terms of Service","url":"https://www.anthropic.com/legal/commercial-terms"}],"url":"https://www.anthropic.com/legal/commercial-terms"},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/anthropic/claude-haiku-5.5","providerModelId":"anthropic/claude-haiku-5.5","verifiedAt":"2026-10-08","sources":[{"label":"Claude Haiku 5.5 on OpenRouter","url":"https://openrouter.ai/anthropic/claude-haiku-5.5"},{"label":"OpenRouter Claude Haiku 5.5 endpoint metadata","url":"https://openrouter.ai/api/v1/models/anthropic/claude-haiku-5.5/endpoints"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"anthropic/claude-haiku-5.5","verifiedAt":"2026-10-08","sources":[{"label":"Claude Haiku 5.5 on OpenRouter","url":"https://openrouter.ai/anthropic/claude-haiku-5.5"},{"label":"OpenRouter Claude Haiku 5.5 endpoint metadata","url":"https://openrouter.ai/api/v1/models/anthropic/claude-haiku-5.5/endpoints"}]},{"kind":"api","label":"Anthropic API","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview","providerModelId":"claude-haiku-5-5","verifiedAt":"2026-10-08","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/claude-haiku-5.5","providerModelId":"anthropic/claude-haiku-5.5","verifiedAt":"2026-10-08","sources":[{"label":"Claude Haiku 5.5 on Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/claude-haiku-5.5"}]}],"pricing":[{"provider":"Anthropic API","providerModelId":"claude-haiku-5-5","observedAt":"2026-10-08","effectiveAt":"2026-10-07","currency":"USD","unit":"per-1M-tokens","input":0.1,"output":0.5,"cachedRead":0.01,"cachedWrite":0.125,"batchInput":0.05,"batchOutput":0.25,"availability":"active","status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"},{"label":"Claude Haiku 5.5 migration guide","url":"https://platform.claude.com/docs/en/models/haiku-5-5/migration-guide"}],"conditions":"Prompts up to 100,000 input tokens. Over 100,000: $0.50/M input, $2.50/M output, $0.05/M cache reads, $0.625/M 5-minute cache writes and $1/M 1-hour cache writes. Short-prompt 1-hour cache writes cost $0.20/M. Batch input and output are 50% off."}],"speed":[{"provider":"Anthropic","observedAt":"2026-10-08","modelSnapshot":"claude-haiku-5-5","reasoningModeOrBudget":"standard speed","status":"verified","sources":[{"label":"Introducing Claude Haiku 5.5","url":"https://www.anthropic.com/claude-haiku-5-5"},{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]}],"related":[{"id":"anthropic/claude-opus-5.5","name":"Claude Opus 5.5","benchmarkUrl":"/benchmarks/models/claude-opus-5.5"},{"id":"anthropic/claude-sonnet-5.5","name":"Claude Sonnet 5.5","benchmarkUrl":"/benchmarks/models/claude-sonnet-5.5"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]},"weightFootprints":[],"componentStorage":{}},"special":"Fast, low-cost Claude with adaptive reasoning and a million-token context. Prompt-length pricing makes short, high-volume work especially inexpensive.","bestUseCase":"Summarization, classification, document lookups, compaction, customer support and focused subagent work alongside larger Claude models.","pricingSummary":{"value":"Anthropic API, USD per 1M tokens: prompts up to 100,000 tokens, input $0.10, output $0.50, cache read $0.01, 5-minute cache write $0.125, 1-hour cache write $0.20; prompts over 100,000 tokens, input $0.50, output $2.50, cache read $0.05, 5-minute cache write $0.625, 1-hour cache write $1.00. Batch input and output are 50% off.","status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]},"speedSummary":{"value":"Anthropic comparative latency: Fastest at standard speed, observed 2026-10-08. No absolute output-token throughput or local hardware measurement established.","status":"verified","sources":[{"label":"Claude Haiku 5.5","url":"https://platform.claude.com/docs/en/models/haiku-5-5/overview"}]},"relatedArtifacts":[],"normalTokensPerSecond":402.19,"normalTokensPerSecondMin":385.78,"normalTokensPerSecondMax":434.93},"google/gemma-4-31b-it":{"id":"google/gemma-4-31b-it","name":"Gemma 4 31B","facts":{"family":{"value":"Gemma 4","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}]},"developer":{"value":"Google DeepMind","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}]},"origin":{"value":"United States","status":"inferred","sources":[{"label":"Gemma 4 release announcement","url":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/"}]},"releasedAt":{"value":"2026-04-02","status":"verified","sources":[{"label":"Gemma 4 release announcement","url":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/"}]},"license":{"value":"Apache 2.0","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}],"url":"https://ai.google.dev/gemma/apache_2"},"architecture":{"value":"Dense multimodal transformer with interleaved sliding-window and global attention","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}]},"totalParameters":{"value":30700000000,"status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]},"contextWindowTokens":{"value":262144,"status":"verified","sources":[{"label":"OpenRouter model catalog","url":"https://openrouter.ai/api/v1/models"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"},{"label":"Gemma 4 release announcement","url":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/"}]},"activeParameters":{"value":30700000000,"status":"inferred","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/google/gemma-4-31b-it","providerModelId":"google/gemma-4-31b-it","verifiedAt":"2026-10-09","sources":[{"label":"Gemma 4 31b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-31b-it/endpoints"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"google/gemma-4-31b-it","verifiedAt":"2026-10-09","sources":[{"label":"Gemma 4 31b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-31b-it/endpoints"}]},{"kind":"router","label":"OpenRouter free model page","url":"https://openrouter.ai/google/gemma-4-31b-it:free","providerModelId":"google/gemma-4-31b-it:free","verifiedAt":"2026-10-09","sources":[{"label":"Google: Gemma 4 31B (free) free endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-31b-it:free/endpoints"}]},{"kind":"api","label":"OpenRouter free Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"google/gemma-4-31b-it:free","verifiedAt":"2026-10-09","sources":[{"label":"Google: Gemma 4 31B (free) free endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-31b-it:free/endpoints"}]},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/google/gemma-4-31B-it","providerModelId":"google/gemma-4-31B-it","verifiedAt":"2026-10-09","sources":[{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"google/gemma-4-31b-it","observedAt":"2026-10-09T15:01:26.665Z","currency":"USD","unit":"per-1M-tokens","input":0.14,"output":0.39999999999999997,"availability":"Representative listed paid provider rate; provider rates vary.","status":"verified","sources":[{"label":"Gemma 4 31b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-31b-it/endpoints"}]},{"provider":"OpenRouter","providerModelId":"google/gemma-4-31b-it:free","observedAt":"2026-10-09T14:36:52.753Z","currency":"USD","unit":"per-1M-tokens","input":0,"output":0,"availability":"Free model route listed by the provider on the observation date.","status":"verified","sources":[{"label":"Google: Gemma 4 31B (free) free endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-31b-it:free/endpoints"}]},{"provider":"OpenRouter","providerModelId":"google/gemma-4-31b-it","observedAt":"2026-10-09T14:10:55.907Z","currency":"USD","unit":"per-1M-tokens","input":0.09,"output":0.33999999999999997,"availability":"From catalog minimum rates; endpoint rates vary.","status":"verified","sources":[{"label":"OpenRouter model catalog","url":"https://openrouter.ai/api/v1/models"}]}],"speed":[],"related":[{"id":"google/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B","benchmarkUrl":"/benchmarks/models/gemma-4-26b-a4b-it"}],"deployment":{"weightFootprints":[],"componentStorage":{}},"special":"Google DeepMind’s open-weight dense model with configurable thinking, image understanding and native function calling.","bestUseCase":"Self-hosted coding assistants, document analysis and agents that need open weights and a long context window.","pricingSummary":{"value":"OpenRouter: $0.14 input / $0.40 output per 1M tokens, observed 2026-10-09; provider rates vary.","status":"verified","sources":[{"label":"Gemma 4 31b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-31b-it/endpoints"}]},"relatedArtifacts":[],"normalTokensPerSecond":116.26,"normalTokensPerSecondMin":75.93,"normalTokensPerSecondMax":158.59},"google/gemma-4-26b-a4b-it":{"id":"google/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B","facts":{"family":{"value":"Gemma 4","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}]},"developer":{"value":"Google DeepMind","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}]},"origin":{"value":"United States","status":"inferred","sources":[{"label":"Gemma 4 release announcement","url":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/"}]},"releasedAt":{"value":"2026-04-02","status":"verified","sources":[{"label":"Gemma 4 release announcement","url":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/"}]},"license":{"value":"Apache 2.0","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 31B instruction model","url":"https://huggingface.co/google/gemma-4-31B-it"}],"url":"https://ai.google.dev/gemma/apache_2"},"architecture":{"value":"Mixture-of-Experts multimodal transformer with interleaved sliding-window and global attention","status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"},{"label":"Official Gemma 4 26B A4B instruction model","url":"https://huggingface.co/google/gemma-4-26B-A4B-it"}]},"totalParameters":{"value":25200000000,"status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]},"contextWindowTokens":{"value":262144,"status":"verified","sources":[{"label":"OpenRouter model catalog","url":"https://openrouter.ai/api/v1/models"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Official Gemma 4 26B A4B instruction model","url":"https://huggingface.co/google/gemma-4-26B-A4B-it"},{"label":"Gemma 4 release announcement","url":"https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/"}]},"activeParameters":{"value":3800000000,"status":"verified","sources":[{"label":"Gemma 4 model card","url":"https://ai.google.dev/gemma/docs/core/model_card_4"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/google/gemma-4-26b-a4b-it","providerModelId":"google/gemma-4-26b-a4b-it","verifiedAt":"2026-10-09","sources":[{"label":"Gemma 4 26b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-26b-a4b-it/endpoints"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"google/gemma-4-26b-a4b-it","verifiedAt":"2026-10-09","sources":[{"label":"Gemma 4 26b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-26b-a4b-it/endpoints"}]},{"kind":"router","label":"OpenRouter free model page","url":"https://openrouter.ai/google/gemma-4-26b-a4b-it:free","providerModelId":"google/gemma-4-26b-a4b-it:free","verifiedAt":"2026-10-09","sources":[{"label":"Google: Gemma 4 26B A4B  (free) free endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-26b-a4b-it:free/endpoints"}]},{"kind":"api","label":"OpenRouter free Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"google/gemma-4-26b-a4b-it:free","verifiedAt":"2026-10-09","sources":[{"label":"Google: Gemma 4 26B A4B  (free) free endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-26b-a4b-it:free/endpoints"}]},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/google/gemma-4-26B-A4B-it","providerModelId":"google/gemma-4-26B-A4B-it","verifiedAt":"2026-10-09","sources":[{"label":"Official Gemma 4 26B A4B instruction model","url":"https://huggingface.co/google/gemma-4-26B-A4B-it"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"google/gemma-4-26b-a4b-it","observedAt":"2026-10-09T15:01:26.729Z","currency":"USD","unit":"per-1M-tokens","input":0.09,"output":0.3,"availability":"Representative listed paid provider rate; provider rates vary.","status":"verified","sources":[{"label":"Gemma 4 26b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-26b-a4b-it/endpoints"}]},{"provider":"OpenRouter","providerModelId":"google/gemma-4-26b-a4b-it:free","observedAt":"2026-10-09T14:36:52.753Z","currency":"USD","unit":"per-1M-tokens","input":0,"output":0,"availability":"Free model route listed by the provider on the observation date.","status":"verified","sources":[{"label":"Google: Gemma 4 26B A4B  (free) free endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-26b-a4b-it:free/endpoints"}]}],"speed":[],"related":[{"id":"google/gemma-4-31b-it","name":"Gemma 4 31B","benchmarkUrl":"/benchmarks/models/gemma-4-31b-it"}],"deployment":{"weightFootprints":[],"componentStorage":{}},"special":"Google DeepMind’s open-weight mixture-of-experts model activates 3.8B parameters per token, with configurable thinking, image understanding and native function calling.","bestUseCase":"Self-hosted coding assistants and multimodal agents where inference efficiency and open weights matter.","pricingSummary":{"value":"OpenRouter: $0.09 input / $0.30 output per 1M tokens, observed 2026-10-09; provider rates vary.","status":"verified","sources":[{"label":"Gemma 4 26b paid endpoint catalog","url":"https://openrouter.ai/api/v1/models/google/gemma-4-26b-a4b-it/endpoints"}]},"relatedArtifacts":[],"normalTokensPerSecond":95,"normalTokensPerSecondMin":94.76,"normalTokensPerSecondMax":95.34},"meta/muse-wonder":{"id":"meta/muse-wonder","name":"Muse Wonder","facts":{"family":{"value":"Muse Wonder","status":"verified","sources":[{"label":"Meta model specifications and Muse Wonder availability","url":"https://dev.meta.ai/docs/models"},{"label":"Muse Wonder release announcement","url":"https://research.meta.ai/blog/muse-wonder"}]},"developer":{"value":"Meta","status":"verified","sources":[{"label":"Meta model specifications and Muse Wonder availability","url":"https://dev.meta.ai/docs/models"},{"label":"Muse Wonder release announcement","url":"https://research.meta.ai/blog/muse-wonder"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Moving to a Local Selling Model","url":"https://about.fb.com/news/2017/12/moving-to-a-local-selling-model/"}]},"releasedAt":{"value":"2026-10-12","status":"verified","sources":[{"label":"Muse Wonder release announcement","url":"https://research.meta.ai/blog/muse-wonder"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Meta model specifications and Muse Wonder availability","url":"https://dev.meta.ai/docs/models"},{"label":"Meta Model API Terms of Service","url":"https://dev.meta.ai/legal/terms-of-service"}],"url":"https://dev.meta.ai/legal/terms-of-service"},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Meta model specifications and Muse Wonder availability","url":"https://dev.meta.ai/docs/models"},{"label":"Muse Wonder OpenRouter model metadata","url":"https://openrouter.ai/api/v1/model/meta/muse-wonder"}]},"inputModalities":{"value":["text","image","video","audio","PDF"],"status":"verified","sources":[{"label":"Meta model specifications and Muse Wonder availability","url":"https://dev.meta.ai/docs/models"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Meta model specifications and Muse Wonder availability","url":"https://dev.meta.ai/docs/models"}]},"weightsAvailability":{"value":"hosted-only","status":"reviewed","sources":[{"label":"Meta Model API overview: Muse Wonder preview and deployment","url":"https://dev.meta.ai/docs/overview"},{"label":"Meta models: Wonder, Spark and their workload comparison","url":"https://dev.meta.ai/docs/models"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/meta/muse-wonder","providerModelId":"meta/muse-wonder","verifiedAt":"2026-10-12","sources":[{"label":"Muse Wonder OpenRouter endpoint catalog","url":"https://openrouter.ai/api/v1/models/meta/muse-wonder/endpoints"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"meta/muse-wonder","verifiedAt":"2026-10-12","sources":[{"label":"Muse Wonder OpenRouter endpoint catalog","url":"https://openrouter.ai/api/v1/models/meta/muse-wonder/endpoints"}]},{"kind":"api","label":"Meta Model API","url":"https://dev.meta.ai/docs/models#muse-wonder","providerModelId":"muse-wonder-1.0","verifiedAt":"2026-10-12","sources":[{"label":"Meta model specifications and Muse Wonder availability","url":"https://dev.meta.ai/docs/models"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/muse-wonder-1.0","providerModelId":"meta/muse-wonder-1.0","verifiedAt":"2026-10-12","sources":[{"label":"Vercel AI Gateway exact Muse Wonder route","url":"https://ai-gateway.vercel.sh/v1/models"}]}],"pricing":[{"provider":"Vercel AI Gateway","providerModelId":"meta/muse-wonder-1.0","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.19999999999999998,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},{"provider":"OpenRouter","providerModelId":"meta/muse-wonder","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.19999999999999998,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Meta Model API","providerModelId":"muse-wonder-1.0","observedAt":"2026-10-12","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.2,"availability":"active","status":"verified","sources":[{"label":"Meta Model API pricing and rate limits","url":"https://dev.meta.ai/docs/pricing-rate-limits"}]},{"provider":"OpenRouter","providerModelId":"meta/muse-wonder","observedAt":"2026-10-12","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.2,"availability":"Historical rates","status":"verified","sources":[{"label":"Muse Wonder OpenRouter endpoint catalog","url":"https://openrouter.ai/api/v1/models/meta/muse-wonder/endpoints"}]}],"speed":[],"related":[{"id":"meta/muse-spark-1.3","name":"Muse Spark 1.3","benchmarkUrl":"/benchmarks/models/muse-spark-1.3"}],"deployment":{"selfHostable":{"value":false,"status":"reviewed","sources":[{"label":"Meta Model API overview: Muse Wonder preview and deployment","url":"https://dev.meta.ai/docs/overview"},{"label":"Meta models: Wonder, Spark and their workload comparison","url":"https://dev.meta.ai/docs/models"}]},"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Meta Model API overview: Muse Wonder preview and deployment","url":"https://dev.meta.ai/docs/overview"},{"label":"Meta models: Wonder, Spark and their workload comparison","url":"https://dev.meta.ai/docs/models"}]},"weightFootprints":[],"componentStorage":{}},"special":"Meta's reasoning model combines a million-token context with image, video, audio and PDF understanding, tool use and computer use.","bestUseCase":"Complex software engineering and long-running agents that need to plan, work across files and recover from errors.","pricingSummary":{"value":"Meta Model API / OpenRouter; input 4 USD/1M; cached read 0.20/1M; output 20/1M","status":"verified","sources":[{"label":"Meta Model API pricing and rate limits","url":"https://dev.meta.ai/docs/pricing-rate-limits"},{"label":"Muse Wonder OpenRouter endpoint catalog","url":"https://openrouter.ai/api/v1/models/meta/muse-wonder/endpoints"}]},"relatedArtifacts":[],"normalTokensPerSecond":570.16,"normalTokensPerSecondMin":413.28,"normalTokensPerSecondMax":672.65},"google/gemini-4-argon":{"id":"google/gemini-4-argon","name":"Gemini 4 Argon","facts":{"family":{"value":"Gemini 4","status":"verified","sources":[{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"},{"label":"Introducing Gemini 4 Argon","url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/"}]},"developer":{"value":"Google","status":"verified","sources":[{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"},{"label":"Introducing Gemini 4 Argon","url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Google Terms of Service","url":"https://www.about.google/policies/terms/"}]},"releasedAt":{"value":"2026-10-13","status":"verified","sources":[{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"},{"label":"Introducing Gemini 4 Argon","url":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Gemini API Additional Terms of Service","url":"https://ai.google.dev/gemini-api/terms?hl=en"},{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"}],"url":"https://ai.google.dev/gemini-api/terms"},"architecture":{"value":"Sparse mixture-of-experts (MoE) transformer with native multimodal inputs","status":"verified","sources":[{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"}]},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Gemini 4 Argon API model specifications","url":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon?hl=en"},{"label":"Gemini 4 Argon OpenRouter endpoint specifications and pricing","url":"https://openrouter.ai/api/v1/models/google/gemini-4-argon/endpoints"}]},"maxOutputTokens":{"value":1048576,"status":"verified","sources":[{"label":"Gemini 4 Argon API model specifications","url":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon?hl=en"},{"label":"OpenRouter model catalog maximum-output pattern","url":"https://openrouter.ai/api/v1/models"}]},"inputModalities":{"value":["text","image","audio","video","PDF"],"status":"verified","sources":[{"label":"Gemini 4 Argon API model specifications","url":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon?hl=en"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Gemini 4 Argon API model specifications","url":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon?hl=en"}]},"weightsAvailability":{"value":"hosted-only","status":"reviewed","sources":[{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/google/gemini-4-argon","providerModelId":"google/gemini-4-argon","verifiedAt":"2026-10-13","sources":[{"label":"Gemini 4 Argon OpenRouter endpoint specifications and pricing","url":"https://openrouter.ai/api/v1/models/google/gemini-4-argon/endpoints"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"google/gemini-4-argon","verifiedAt":"2026-10-13","sources":[{"label":"Gemini 4 Argon OpenRouter endpoint specifications and pricing","url":"https://openrouter.ai/api/v1/models/google/gemini-4-argon/endpoints"}]},{"kind":"api","label":"Gemini Developer API","url":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon","providerModelId":"gemini-4-argon","verifiedAt":"2026-10-13","sources":[{"label":"Gemini 4 Argon API model specifications","url":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon?hl=en"}]},{"kind":"gateway","label":"Google AI Studio","url":"https://aistudio.google.com/prompts/new_chat?model=gemini-4-argon","providerModelId":"gemini-4-argon","verifiedAt":"2026-10-13","sources":[{"label":"Gemini 4 Argon API model specifications","url":"https://ai.google.dev/gemini-api/docs/models/gemini-4-argon?hl=en"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"google/gemini-4-argon","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.09999999999999999,"cachedWrite":0.1875,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Gemini Developer API","providerModelId":"gemini-4-argon","observedAt":"2026-10-13","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.1,"availability":"active","status":"verified","sources":[{"label":"Gemini Developer API pricing","url":"https://ai.google.dev/gemini-api/docs/pricing.md"}],"conditions":"Standard introductory price through 30 November 2026; output includes thinking; implicit caching only."},{"provider":"Gemini Developer API","providerModelId":"gemini-4-argon","observedAt":"2026-10-13","effectiveAt":"2026-12-01","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.2,"availability":"announced-future-rate","status":"verified","sources":[{"label":"Gemini Developer API pricing","url":"https://ai.google.dev/gemini-api/docs/pricing.md"}],"conditions":"Announced Standard price effective 1 December 2026."},{"provider":"Gemini Developer API / Flex","providerModelId":"gemini-4-argon","observedAt":"2026-10-13","currency":"USD","unit":"per-1M-tokens","input":1,"output":5,"cachedRead":0.05,"availability":"active","status":"verified","sources":[{"label":"Gemini Developer API pricing","url":"https://ai.google.dev/gemini-api/docs/pricing.md"}],"conditions":"Flex introductory price through 30 November 2026; separate service tier."}],"speed":[],"related":[{"id":"google/gemini-3.8-flash","name":"Gemini 3.8 Flash","benchmarkUrl":"/benchmarks/models/gemini-3.8-flash"}],"deployment":{"selfHostable":{"value":false,"status":"reviewed","sources":[{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"}]},"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Gemini 4 Argon model card","url":"https://deepmind.google/models/model-cards/gemini-4-argon/"}]},"weightFootprints":[],"componentStorage":{}},"special":"Google’s multimodal reasoning model combines sparse mixture-of-experts architecture with sustained reasoning and a shared million-token input/output context.","bestUseCase":"Complex software engineering, scientific and mathematical problem solving, and long-running agents working across documents, images, audio and video.","pricingSummary":{"value":"OpenRouter / Gemini API Standard: $2 input, $0.10 cached input and $10 output per 1M tokens through 2026-11-30; Gemini API Standard becomes $4 / $0.20 / $20 on 2026-12-01. Output includes thinking tokens. Flex is a separate tier at $1 input / $0.05 cached input / $5 output during the introductory period.","status":"verified","sources":[{"label":"Gemini Developer API pricing","url":"https://ai.google.dev/gemini-api/docs/pricing.md"},{"label":"Gemini 4 Argon OpenRouter endpoint specifications and pricing","url":"https://openrouter.ai/api/v1/models/google/gemini-4-argon/endpoints"}]},"relatedArtifacts":[],"normalTokensPerSecond":143.66,"normalTokensPerSecondMin":138.69,"normalTokensPerSecondMax":147.61},"meta/muse-spark-open":{"id":"meta/muse-spark-open","name":"Muse Spark Open","facts":{"family":{"value":"Muse Spark","status":"verified","sources":[{"label":"Muse Spark Open benchmark table","url":"https://dev.meta.ai/models/muse-spark-open"}]},"developer":{"value":"Meta","status":"verified","sources":[{"label":"Muse Spark Open launch","url":"https://research.meta.ai/blog/muse-spark-open"}]},"origin":{"value":"United States","status":"reviewed","sources":[{"label":"Moving to a Local Selling Model","url":"https://about.fb.com/news/2017/12/moving-to-a-local-selling-model/"}]},"releasedAt":{"value":"2026-10-15","status":"verified","sources":[{"label":"Muse Spark Open launch","url":"https://research.meta.ai/blog/muse-spark-open"}]},"license":{"value":"Muse Model License","status":"verified","sources":[{"label":"Muse Model License","url":"https://huggingface.co/unsloth/Muse-Spark-Open-816B-A42B-GGUF/raw/main/LICENSE.md"}],"url":"https://huggingface.co/unsloth/Muse-Spark-Open-816B-A42B-GGUF/blob/main/LICENSE.md"},"architecture":{"value":"Sparse mixture-of-experts Transformer with perception encoders; 62 decoder layers, 256 experts with 8 selected per token, grouped-query attention and sliding-window/full-attention layers.","status":"verified","sources":[{"label":"Meta model card reproduced by Unsloth","url":"https://huggingface.co/unsloth/Muse-Spark-Open-816B-A42B/raw/main/README.md"},{"label":"Recommended settings","url":"https://dev.meta.ai/docs/muse-spark-open/settings"}]},"totalParameters":{"value":816000000000,"status":"verified","sources":[{"label":"Muse Spark Open overview","url":"https://dev.meta.ai/docs/muse-spark-open"}]},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Muse Spark Open overview","url":"https://dev.meta.ai/docs/muse-spark-open"},{"label":"Vercel route metadata and pricing","url":"https://ai-gateway.vercel.sh/v1/models"}]},"maxOutputTokens":{"value":1048576,"status":"verified","sources":[{"label":"Vercel route metadata and pricing","url":"https://ai-gateway.vercel.sh/v1/models"}]},"inputModalities":{"value":["text","image","audio","video"],"status":"verified","sources":[{"label":"Muse Spark Open overview","url":"https://dev.meta.ai/docs/muse-spark-open"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Muse Spark Open overview","url":"https://dev.meta.ai/docs/muse-spark-open"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Unsloth GGUF repository metadata","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF"}]},"activeParameters":{"value":42000000000,"status":"verified","sources":[{"label":"Muse Spark Open overview","url":"https://dev.meta.ai/docs/muse-spark-open"}]}},"access":[{"kind":"gguf","label":"Unsloth GGUF downloads","url":"https://huggingface.co/unsloth/Muse-Spark-Open-816B-A42B-GGUF","providerModelId":"unsloth/Muse-Spark-Open-816B-A42B-GGUF","verifiedAt":"2026-10-15","sources":[{"label":"Unsloth GGUF repository metadata","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/muse-spark-open","providerModelId":"meta/muse-spark-open","verifiedAt":"2026-10-15","sources":[{"label":"Vercel route metadata and pricing","url":"https://ai-gateway.vercel.sh/v1/models"}]},{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/meta/muse-spark-open","providerModelId":"meta/muse-spark-open","verifiedAt":"2026-10-15","sources":[{"label":"OpenRouter exact route metadata","url":"https://openrouter.ai/api/v1/models"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"meta/muse-spark-open","verifiedAt":"2026-10-15","sources":[{"label":"OpenRouter exact route metadata","url":"https://openrouter.ai/api/v1/models"}]}],"pricing":[{"provider":"Vercel AI Gateway","providerModelId":"meta/muse-spark-open","currency":"USD","unit":"per-1M-tokens","input":1.25,"output":4.25,"cachedRead":0.15,"fetchedAt":"2026-10-15T21:13:27.413Z","observedAt":"2026-10-15T21:13:27.413Z","status":"verified","sources":[{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},{"provider":"OpenRouter","providerModelId":"meta/muse-spark-open","currency":"USD","unit":"per-1M-tokens","input":1.25,"output":4.25,"cachedRead":0.15,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]}],"speed":[],"related":[{"id":"meta/muse-spark-1.3","name":"Muse Spark 1.3","benchmarkUrl":"/benchmarks/models/muse-spark-1.3"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Unsloth GGUF repository metadata","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF"}]},"deploymentClass":{"value":"High-memory local or multi-GPU server; Unsloth quantization and CPU/GPU offload supported","status":"verified","sources":[{"label":"Muse Spark Open local deployment guide","url":"https://unsloth.ai/docs/models/muse-spark-open"}]},"weightFootprints":[{"artifact":"UD-IQ1_M","precisionOrQuantization":"UD-IQ1_M","sizeGB":208.925571064,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]},{"artifact":"UD-IQ1_S","precisionOrQuantization":"UD-IQ1_S","sizeGB":175.639574392,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]},{"artifact":"UD-IQ2_S","precisionOrQuantization":"UD-IQ2_S","sizeGB":275.49756428,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]},{"artifact":"UD-IQ2_XXS","precisionOrQuantization":"UD-IQ2_XXS","sizeGB":225.568569304,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]},{"artifact":"UD-IQ3_XXS","precisionOrQuantization":"UD-IQ3_XXS","sizeGB":340.350896376,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]},{"artifact":"UD-Q2_K_XL","precisionOrQuantization":"UD-Q2_K_XL","sizeGB":311.225649368,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]},{"artifact":"UD-Q3_K_XL","precisionOrQuantization":"UD-Q3_K_XL","sizeGB":401.483491832,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]},{"artifact":"UD-Q4_K_XL","precisionOrQuantization":"UD-Q4_K_XL","sizeGB":505.502231288,"status":"artifact-derived","sources":[{"label":"Unsloth GGUF file sizes","url":"https://huggingface.co/api/models/unsloth/Muse-Spark-Open-816B-A42B-GGUF/tree/main?recursive=true&expand=false"}]}],"componentStorage":{}},"special":"Meta's open-weight coding and agent model combines 816B total parameters, 42B active per token and native image, audio and video understanding.","bestUseCase":"Coding agents, repository analysis and multimodal workflows, with hosted APIs or Unsloth GGUFs for local deployment.","pricingSummary":{"value":"Vercel AI Gateway / OpenRouter; input 1.25 USD/1M; cached read 0.15/1M; output 4.25/1M","status":"verified","sources":[{"label":"Vercel route metadata and pricing","url":"https://ai-gateway.vercel.sh/v1/models"},{"label":"OpenRouter exact route metadata","url":"https://openrouter.ai/api/v1/models"},{"label":"Configured Nous route metadata","url":"https://inference-api.nousresearch.com/v1/models"}]},"relatedArtifacts":[],"normalTokensPerSecond":361.08,"normalTokensPerSecondMin":290.98,"normalTokensPerSecondMax":424.52},"reflection/beam-1.0-501b-a23b":{"id":"reflection/beam-1.0-501b-a23b","name":"Beam 1.0 501B A23B","facts":{"family":{"value":"Beam 1.0","status":"verified","sources":[{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"}]},"developer":{"value":"Reflection AI","status":"verified","sources":[{"label":"Beam is now available","url":"https://reflection.ai/blog/beam-is-available"},{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Reflection organization address","url":"https://reflection.ai/legal/european-privacy-policy"}]},"releasedAt":{"value":"2026-10-16","status":"verified","sources":[{"label":"Beam is now available","url":"https://reflection.ai/blog/beam-is-available"}]},"license":{"value":"Apache 2.0","status":"verified","sources":[{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"}],"url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B/blob/main/LICENSE"},"architecture":{"value":"Sparse mixture-of-experts transformer; 52 layers (3 dense, 49 MoE), 256 experts with 8 selected per token; gated GQA and sliding-window attention","status":"verified","sources":[{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"},{"label":"Beam configuration","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B/blob/main/config.json"}]},"totalParameters":{"value":501267612352,"status":"verified","sources":[{"label":"Beam checkpoint parameter metadata","url":"https://huggingface.co/api/models/reflection/Beam-1.0-501B-A23B"}]},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Reflection model specifications","url":"https://developers.reflection.ai/models"},{"label":"OpenRouter Beam endpoints","url":"https://openrouter.ai/api/v1/models/reflection/beam-1.0-501b-a23b/endpoints"}]},"maxOutputTokens":{"value":131072,"status":"verified","sources":[{"label":"Reflection model specifications","url":"https://developers.reflection.ai/models"},{"label":"OpenRouter Beam endpoints","url":"https://openrouter.ai/api/v1/models/reflection/beam-1.0-501b-a23b/endpoints"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Beam is now available","url":"https://reflection.ai/blog/beam-is-available"},{"label":"Official Beam 1.0 weights collection","url":"https://huggingface.co/collections/reflection/beam-10"},{"label":"Beam 1.0 Hugging Face repository metadata","url":"https://huggingface.co/api/models/reflection/Beam-1.0-501B-A23B"}]},"activeParameters":{"value":23400000000,"status":"verified","sources":[{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"}]}},"access":[{"kind":"gateway","label":"OpenCode Beam Free","url":"https://opencode.ai/v2/docs/console/models/","providerModelId":"beam-free","verifiedAt":"2026-10-16","sources":[{"label":"OpenCode models, prices and privacy","url":"https://opencode.ai/v2/docs/console/models/"}]},{"kind":"router","label":"OpenRouter Beam","url":"https://openrouter.ai/reflection/beam-1.0-501b-a23b","providerModelId":"reflection/beam-1.0-501b-a23b","verifiedAt":"2026-10-16","sources":[{"label":"OpenRouter Beam prices and providers","url":"https://openrouter.ai/reflection/beam-1.0-501b-a23b"}]},{"kind":"api","label":"Reflection API quickstart","url":"https://developers.reflection.ai/quickstart","providerModelId":"Beam-1.0-501B-A23B","verifiedAt":"2026-10-16","sources":[{"label":"Reflection API quickstart","url":"https://developers.reflection.ai/quickstart"}]},{"kind":"gateway","label":"Reflection platform","url":"https://platform.reflection.ai/","providerModelId":"Beam-1.0-501B-A23B","verifiedAt":"2026-10-16","sources":[{"label":"Reflection API quickstart","url":"https://developers.reflection.ai/quickstart"}]},{"kind":"download","label":"Official BF16 weights","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B","providerModelId":"reflection/Beam-1.0-501B-A23B","verifiedAt":"2026-10-16","sources":[{"label":"Official Beam 1.0 weights collection","url":"https://huggingface.co/collections/reflection/beam-10"}]},{"kind":"download","label":"Official FP8 weights","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B-FP8","providerModelId":"reflection/Beam-1.0-501B-A23B-FP8","verifiedAt":"2026-10-16","sources":[{"label":"Official Beam 1.0 weights collection","url":"https://huggingface.co/collections/reflection/beam-10"}]},{"kind":"download","label":"Official NVFP4 weights","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B-NVFP4","providerModelId":"reflection/Beam-1.0-501B-A23B-NVFP4","verifiedAt":"2026-10-16","sources":[{"label":"Official Beam 1.0 weights collection","url":"https://huggingface.co/collections/reflection/beam-10"}]},{"kind":"download","label":"Official DFlash2 drafter","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B-DFlash2","providerModelId":"reflection/Beam-1.0-501B-A23B-DFlash2","verifiedAt":"2026-10-16","sources":[{"label":"Official Beam 1.0 weights collection","url":"https://huggingface.co/collections/reflection/beam-10"}]},{"kind":"gguf","label":"Unsloth GGUF weights","url":"https://huggingface.co/unsloth/Beam-1.0-501B-A23B-GGUF","providerModelId":"unsloth/Beam-1.0-501B-A23B-GGUF","verifiedAt":"2026-10-16","sources":[{"label":"Unsloth Beam GGUF conversions","url":"https://huggingface.co/unsloth/Beam-1.0-501B-A23B-GGUF"}]},{"kind":"gateway","label":"Vercel Beam Free","url":"https://vercel.com/ai-gateway/models/beam-free","providerModelId":"reflection/beam-free","verifiedAt":"2026-10-16","sources":[{"label":"Vercel Beam Free listing","url":"https://vercel.com/ai-gateway/models/beam-free"}]}],"pricing":[{"provider":"Vercel AI Gateway","providerModelId":"reflection/beam-free","currency":"USD","unit":"per-1M-tokens","input":0,"output":0,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},{"provider":"OpenRouter","providerModelId":"reflection/beam-1.0-501b-a23b","currency":"USD","unit":"per-1M-tokens","input":0.25,"output":1,"cachedRead":0.049999999999999996,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"OpenCode","providerModelId":"beam-free","observedAt":"2026-10-16","currency":"USD","unit":"per-1M-tokens","input":0,"output":0,"cachedRead":0,"availability":"limited-time promotion","status":"verified","sources":[{"label":"OpenCode models, prices and privacy","url":"https://opencode.ai/v2/docs/console/models/"}],"conditions":"Limited-time free offer; account access and regional availability follow OpenCode’s terms."},{"provider":"OpenRouter / Nebius","providerModelId":"reflection/beam-1.0-501b-a23b","observedAt":"2026-10-16","currency":"USD","unit":"per-1M-tokens","input":0.25,"output":1,"availability":"active","status":"verified","sources":[{"label":"OpenRouter Beam prices and providers","url":"https://openrouter.ai/reflection/beam-1.0-501b-a23b"},{"label":"OpenRouter Beam endpoints","url":"https://openrouter.ai/api/v1/models/reflection/beam-1.0-501b-a23b/endpoints"}],"conditions":"Listed Nebius FP8 endpoint; input and output rates per million tokens."}],"speed":[],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Beam is now available","url":"https://reflection.ai/blog/beam-is-available"},{"label":"Official Beam 1.0 weights collection","url":"https://huggingface.co/collections/reflection/beam-10"}]},"deploymentClass":{"value":"large multi-GPU server or substantial host-memory offload","status":"reviewed","sources":[{"label":"Beam 1.0 model card","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B"},{"label":"Serving Reflection Beam on vLLM from Day 0","url":"https://vllm.ai/blog/2026-10-16-reflection-beam"}]},"weightFootprints":[{"artifact":"Official BF16 checkpoint","precisionOrQuantization":"BF16 with F32 tensors","sizeGB":1002.663700352,"status":"verified","sources":[{"label":"BF16 weight payload index","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B/blob/main/model.safetensors.index.json"}]},{"artifact":"Official FP8 checkpoint","precisionOrQuantization":"FP8","sizeGB":503.276437376,"status":"verified","sources":[{"label":"FP8 weight payload index","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B-FP8/blob/main/model.safetensors.index.json"}]}],"componentStorage":{},"contextMemoryCaveat":{"value":"Weight payload size excludes KV cache, runtime overhead and additional serving components; it is not a minimum VRAM requirement.","status":"reviewed","sources":[{"label":"BF16 weight payload index","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B/blob/main/model.safetensors.index.json"},{"label":"FP8 weight payload index","url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B-FP8/blob/main/model.safetensors.index.json"}]}},"special":"Reflection’s open-weight reasoning model activates 23.4B of its 501.3B parameters per token, with a million-token context for coding and long-running agents.","bestUseCase":"Software engineering, multi-step tool workflows and mathematical or scientific reasoning across large codebases and documents.","pricingSummary":{"value":"OpenCode Beam Free: $0 input / $0 output for a limited time. OpenRouter / Nebius: $0.25 input / $1 output per 1M tokens. Reflection advertises free-to-start public beta access; no exact first-party rate verified.","status":"verified","sources":[{"label":"OpenCode models, prices and privacy","url":"https://opencode.ai/v2/docs/console/models/"},{"label":"OpenRouter Beam prices and providers","url":"https://openrouter.ai/reflection/beam-1.0-501b-a23b"},{"label":"Beam is now available","url":"https://reflection.ai/blog/beam-is-available"}]},"relatedArtifacts":[],"normalTokensPerSecond":972.09,"normalTokensPerSecondMin":883.62,"normalTokensPerSecondMax":1148.11},"deepseek/deepseek-v4.1-flash-1017":{"id":"deepseek/deepseek-v4.1-flash-1017","name":"DeepSeek-V4.1-Flash-1017","facts":{"family":{"value":"DeepSeek V4.1","status":"verified","sources":[{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},"developer":{"value":"DeepSeek-AI","status":"verified","sources":[{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"DeepSeek Privacy Policy","url":"https://cdn.deepseek.com/policies/en-US/deepseek-privacy-policy.html"}]},"releasedAt":{"value":"2026-10-17","status":"verified","sources":[{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},"license":{"value":"MIT","status":"verified","sources":[{"label":"DeepSeek 1017 MIT license","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017/blob/main/LICENSE"}],"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017/blob/main/LICENSE"},"architecture":{"value":"Sparse mixture-of-experts with a causal encoder-decoder design and native image understanding.","status":"verified","sources":[{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},"totalParameters":{"value":763205315794,"status":"reviewed","sources":[{"label":"DeepSeek 1017 checkpoint parameter metadata","url":"https://huggingface.co/api/models/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"DeepSeek 1017 model configuration","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017/blob/main/config.json"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"}]},"maxOutputTokens":{"value":393216,"status":"verified","sources":[{"label":"DeepSeek V4.1 Flash model specifications","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"DeepSeek 1017 MIT license","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017/blob/main/LICENSE"}]},"prefillActiveParameters":{"value":8000000000,"status":"inferred","sources":[{"label":"DeepSeek V4.1 Flash model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/README.md"},{"label":"DeepSeek V4.1 Flash model specifications","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"},{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"DeepSeek 1017 model configuration","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017/blob/main/config.json"}]},"decodeActiveParameters":{"value":16000000000,"status":"inferred","sources":[{"label":"DeepSeek V4.1 Flash model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/raw/main/README.md"},{"label":"DeepSeek V4.1 Flash model specifications","url":"https://www.alibabacloud.com/help/en/model-studio/deepseek-v4-1-flash"},{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"},{"label":"DeepSeek 1017 model configuration","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017/blob/main/config.json"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/deepseek/deepseek-v4.1-flash-1017","providerModelId":"deepseek/deepseek-v4.1-flash-1017","verifiedAt":"2026-10-18","sources":[{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"deepseek/deepseek-v4.1-flash-1017","verifiedAt":"2026-10-18","sources":[{"label":"OpenRouter exact 1017 model metadata","url":"https://openrouter.ai/api/v1/model/deepseek/deepseek-v4.1-flash-1017"}]},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017","providerModelId":"deepseek-ai/DeepSeek-V4.1-Flash-1017","verifiedAt":"2026-10-18","sources":[{"label":"DeepSeek 1017 model card","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"deepseek/deepseek-v4.1-flash-1017","observedAt":"2026-10-18","currency":"USD","unit":"per-1M-tokens","input":0.3,"output":1.2,"cachedRead":0.006,"availability":"active","status":"verified","sources":[{"label":"OpenRouter exact 1017 model feed pricing","url":"https://openrouter.ai/api/v1/models"}],"conditions":"Weekday peak rate. Half price outside 01:00-04:00 and 06:00-10:00 UTC on weekdays."}],"speed":[],"related":[{"id":"deepseek/deepseek-v4.1-flash","name":"DeepSeek-V4.1-Flash","benchmarkUrl":"/benchmarks/models/deepseek-v4.1-flash"}],"deployment":{"weightFootprints":[],"componentStorage":{}},"special":"The 17 October checkpoint improves reasoning, coding, agentic tasks and visual understanding, with shorter outputs and faster task completion than the September release.","bestUseCase":"Coding agents, repository changes and long tasks that combine text and images.","relatedArtifacts":[],"normalTokensPerSecond":310.05,"normalTokensPerSecondMin":226.08,"normalTokensPerSecondMax":424.94},"minimax/minimax-m3.1-flash":{"id":"minimax/minimax-m3.1-flash","name":"MiniMax M3.1 Flash","facts":{"family":{"value":"MiniMax M3","status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"developer":{"value":"MiniMax","status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"license":{"value":"MiniMax Community License","status":"verified","sources":[{"label":"MiniMax Community License (MiniMax-M3.1-Flash)","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash/blob/main/LICENSE"}],"url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash/blob/main/LICENSE"},"architecture":{"value":"Mixture-of-Experts with MiniMax Sparse Attention and native multimodal encoder","status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"totalParameters":{"value":428000000000,"status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"},{"label":"Hugging Face model metadata: MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3.1-Flash"}]},"activeParameters":{"value":23000000000,"status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]}},"access":[{"kind":"router","label":"OpenRouter MiniMax M3.1 Flash (free)","url":"https://openrouter.ai/minimax/minimax-m3.1-flash:free","providerModelId":"minimax/minimax-m3.1-flash:free"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"minimax/minimax-m3.1-flash:free"},{"kind":"api","label":"MiniMax API pricing/model access","url":"https://platform.minimax.io/docs/pricing/overview","providerModelId":"MiniMax-M3.1-Flash"},{"kind":"gateway","label":"OpenCode MiniMax M3.1 Flash Free (limited time)","url":"https://opencode.ai/v2/docs/console/models/","providerModelId":"minimax-m3.1-flash-free"}],"pricing":[],"speed":[],"related":[{"id":"minimax/minimax-m3","name":"MiniMax M3","benchmarkUrl":"/benchmarks/models/minimax-m3"},{"id":"space-bunny-alpha","name":"Space Bunny Alpha (frozen)","guideUrl":"https://spacebunnyalpha.com/","benchmarkUrl":"/benchmarks/models/space-bunny-alpha"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"},{"label":"Hugging Face model metadata: MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3.1-Flash"}]},"deploymentClass":{"value":"multi-GPU server","status":"inferred","sources":[{"label":"MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash"}]},"weightFootprints":[{"artifact":"Official safetensors (mixed precision)","precisionOrQuantization":"mixed (MXFP8 with packed 4-bit experts)","sizeGB":252.4,"status":"verified","sources":[{"label":"Hugging Face model metadata: MiniMaxAI/MiniMax-M3.1-Flash","url":"https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3.1-Flash"}]}],"componentStorage":{}},"special":"MiniMax’s low-cost Flash model for agentic work. It ran anonymously as Space Bunny Alpha before its 19 October 2026 release, and MiniMax calls the stealth model an early version.","bestUseCase":"High-volume agentic and coding workloads through the MiniMax API, M Plan or MiniMax Code, where flat low token prices matter.","relatedArtifacts":[]},"minimax/minimax-m3":{"id":"minimax/minimax-m3","name":"MiniMax M3","facts":{"family":{"value":"MiniMax M3","status":"verified","sources":[{"label":"MiniMax M3: Frontier Coding, 1M Context, Native Multimodality","url":"https://www.minimax.io/blog/minimax-m3"},{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"},{"label":"MiniMax M3 Community License","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE"}]},"developer":{"value":"MiniMax","status":"verified","sources":[{"label":"MiniMax M3: Frontier Coding, 1M Context, Native Multimodality","url":"https://www.minimax.io/blog/minimax-m3"},{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"},{"label":"MiniMax M3 Community License","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE"}]},"origin":{"value":"Singapore (MiniMax API service entity Nanonoble Pte. Ltd.)","status":"verified","sources":[{"label":"MiniMax API Privacy Policy","url":"https://platform.minimax.io/protocol/privacy-policy"}]},"releasedAt":{"value":"2026-06-01","status":"verified","sources":[{"label":"MiniMax M3: Frontier Coding, 1M Context, Native Multimodality","url":"https://www.minimax.io/blog/minimax-m3"}]},"license":{"value":"MiniMax Community License","status":"reviewed","sources":[{"label":"MiniMax M3 Community License","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE"}],"url":"https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE"},"architecture":{"value":"Mixture-of-Experts with MiniMax Sparse Attention and native multimodal encoder","status":"verified","sources":[{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"}]},"totalParameters":{"value":428000000000,"status":"verified","sources":[{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"MiniMax M3: Frontier Coding, 1M Context, Native Multimodality","url":"https://www.minimax.io/blog/minimax-m3"},{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"}]},"maxOutputTokens":{"value":524288,"status":"verified","sources":[{"label":"MiniMax Chat Completions API reference","url":"https://platform.minimax.io/docs/api-reference/text-chat-openai"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"MiniMax M3: Frontier Coding, 1M Context, Native Multimodality","url":"https://www.minimax.io/blog/minimax-m3"},{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"MiniMax M3: Frontier Coding, 1M Context, Native Multimodality","url":"https://www.minimax.io/blog/minimax-m3"},{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"MiniMax M3 Community License","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE"}]},"activeParameters":{"value":23000000000,"status":"verified","sources":[{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/minimax/minimax-m3","providerModelId":"minimax/minimax-m3"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"minimax/minimax-m3"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3","providerModelId":"MiniMaxAI/MiniMax-M3","verifiedAt":"2026-10-06","sources":[{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"},{"label":"MiniMax M3 Community License","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE"}]},{"kind":"gguf","label":"Unsloth GGUF (experimental)","url":"https://huggingface.co/unsloth/MiniMax-M3-GGUF","providerModelId":"unsloth/MiniMax-M3-GGUF","verifiedAt":"2026-10-06","sources":[{"label":"MiniMax-M3-GGUF","url":"https://huggingface.co/unsloth/MiniMax-M3-GGUF"}]},{"kind":"api","label":"MiniMax API pricing/model access","url":"https://platform.minimax.io/docs/pricing/overview","providerModelId":"MiniMax-M3","verifiedAt":"2026-10-06","sources":[{"label":"MiniMax API Pricing Overview","url":"https://platform.minimax.io/docs/pricing/overview"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"minimax/minimax-m3","currency":"USD","unit":"per-1M-tokens","input":0.3,"output":1.2,"cachedRead":0.06,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"MiniMax API","providerModelId":"MiniMax-M3","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.3,"output":1.2,"cachedRead":0.06,"availability":"active","status":"verified","sources":[{"label":"MiniMax API Pricing Overview","url":"https://platform.minimax.io/docs/pricing/overview"}]}],"speed":[{"provider":"MiniMax","observedAt":"2026-10-06","status":"verified","sources":[{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":100.4,"timeToFirstTokenMs":2110,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"MiniMax-M3 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/minimax-m3"}]}],"related":[{"id":"minimax/minimax-m3.1-flash","name":"MiniMax M3.1 Flash","guideUrl":"https://spacebunnyalpha.com/","benchmarkUrl":"/benchmarks/models/minimax-m3.1-flash"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"},{"label":"MiniMax-M3-GGUF","url":"https://huggingface.co/unsloth/MiniMax-M3-GGUF"}]},"deploymentClass":{"value":"multi-GPU server","status":"inferred","sources":[{"label":"MiniMax-M3","url":"https://huggingface.co/MiniMaxAI/MiniMax-M3"},{"label":"MiniMax-M3-GGUF","url":"https://huggingface.co/unsloth/MiniMax-M3-GGUF"}]},"weightFootprints":[{"artifact":"Unsloth UD-IQ1_M","precisionOrQuantization":"~1-bit","sizeGB":128,"status":"verified","sources":[{"label":"MiniMax-M3-GGUF","url":"https://huggingface.co/unsloth/MiniMax-M3-GGUF"}]},{"artifact":"Unsloth BF16","precisionOrQuantization":"BF16","sizeGB":852,"status":"verified","sources":[{"label":"MiniMax-M3-GGUF","url":"https://huggingface.co/unsloth/MiniMax-M3-GGUF"}]}],"componentStorage":{}},"special":"Open-weight multimodal MoE with 428B total parameters, 23B active and a million-token context, under MiniMax's community license.","bestUseCase":"Long-context multimodal agents through a low-cost API, or self-hosting on large servers under the model's license terms.","pricingSummary":{"value":"MiniMax API; input 0.3 USD/1M; cached read 0.06/1M; output 1.2/1M; Displayed permanent 50% discount: <=512K $0.30/$1.20; 512K-1M $0.60/$2.40. List rates are twice those values.","status":"verified","sources":[{"label":"MiniMax API Pricing Overview","url":"https://platform.minimax.io/docs/pricing/overview"}]},"speedSummary":{"value":"Artificial Analysis: 100.4 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"MiniMax-M3 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/minimax-m3"}]},"relatedArtifacts":[],"normalTokensPerSecond":161.72,"normalTokensPerSecondMin":133.6,"normalTokensPerSecondMax":195.82},"xiaomi/mimo-v2.6-pro":{"id":"xiaomi/mimo-v2.6-pro","name":"MiMo-V2.6-Pro","facts":{"family":{"value":"MiMo V2.6","status":"verified","sources":[{"label":"MiMo V2.6 release and open-source notice","url":"https://mimo.mi.com/docs/en-US/news/latest/v2-6"},{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"}]},"developer":{"value":"Xiaomi","status":"verified","sources":[{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"},{"label":"MiMo V2.6 release and open-source notice","url":"https://mimo.mi.com/docs/en-US/news/latest/v2-6"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"Xiaomi Launch May 2026 Terms and Conditions","url":"https://www.mi.com/global/support/terms/xiaomi-launch-may-2026/"}]},"releasedAt":{"value":"2026-09-21","status":"reviewed","sources":[{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"MiMo V2.6 release and open-source notice","url":"https://mimo.mi.com/docs/en-US/news/latest/v2-6"}]},"license":{"value":"MIT","status":"reviewed","sources":[{"label":"MiMo-V2.6-Pro-RL model card","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL"},{"label":"MiMo-V2.6-Pro-RL checkpoint parameter metadata","url":"https://huggingface.co/api/models/XiaomiMiMo/MiMo-V2.6-Pro-RL"}],"url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL"},"architecture":{"value":"Sparse Mixture-of-Experts; OpenRouter maps the route to the released Pro-RL checkpoint. The UltraSpeed edition is a separate route","status":"reviewed","sources":[{"label":"MiMo-V2.6-Pro-RL model card","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL"},{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"MiMo V2.6 release and open-source notice","url":"https://mimo.mi.com/docs/en-US/news/latest/v2-6"}]},"totalParameters":{"value":1020000000000,"status":"reviewed","sources":[{"label":"MiMo-V2.6-Pro-RL model card","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL"},{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"}]},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"OpenRouter MiMo V2.6 Pro endpoints","url":"https://openrouter.ai/api/v1/models/xiaomi/mimo-v2.6-pro/endpoints"},{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"}]},"maxOutputTokens":{"value":131072,"status":"verified","sources":[{"label":"OpenRouter MiMo V2.6 Pro endpoints","url":"https://openrouter.ai/api/v1/models/xiaomi/mimo-v2.6-pro/endpoints"},{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"}]},"inputModalities":{"value":["text","image","video","audio"],"status":"verified","sources":[{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"},{"label":"OpenRouter MiMo V2.6 Pro endpoints","url":"https://openrouter.ai/api/v1/models/xiaomi/mimo-v2.6-pro/endpoints"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"},{"label":"OpenRouter MiMo V2.6 Pro endpoints","url":"https://openrouter.ai/api/v1/models/xiaomi/mimo-v2.6-pro/endpoints"}]},"weightsAvailability":{"value":"official-open-variants","status":"reviewed","sources":[{"label":"MiMo-V2.6-Pro-RL model card","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL"},{"label":"MiMo-V2.6-Pro-RL checkpoint parameter metadata","url":"https://huggingface.co/api/models/XiaomiMiMo/MiMo-V2.6-Pro-RL"}]},"activeParameters":{"value":42000000000,"status":"reviewed","sources":[{"label":"MiMo-V2.6-Pro-RL model card","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/xiaomi/mimo-v2.6-pro","providerModelId":"xiaomi/mimo-v2.6-pro"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"xiaomi/mimo-v2.6-pro"},{"kind":"download","label":"Official MiMo V2.6 Pro RL weights","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL","providerModelId":"XiaomiMiMo/MiMo-V2.6-Pro-RL"}],"pricing":[{"provider":"OpenRouter","providerModelId":"xiaomi/mimo-v2.6-pro","currency":"USD","unit":"per-1M-tokens","input":0.435,"output":0.87,"cachedRead":0.0036,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Xiaomi MiMo API","providerModelId":"mimo-v2.6-pro","observedAt":"2026-10-23","currency":"USD","unit":"per-1M-tokens","input":0.435,"output":0.87,"cachedRead":0.0036,"availability":"active","status":"verified","sources":[{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"},{"label":"OpenRouter MiMo V2.6 Pro endpoints","url":"https://openrouter.ai/api/v1/models/xiaomi/mimo-v2.6-pro/endpoints"}]}],"speed":[],"related":[{"id":"xiaomi/mimo-v2.6-flash","name":"MiMo-V2.6-Flash","benchmarkUrl":"/benchmarks/models/mimo-v2.6-flash"}],"deployment":{"weightFootprints":[],"componentStorage":{}},"special":"Xiaomi's flagship MiMo V2.6 multimodal reasoning model, with a million-token context and 128K output. Xiaomi publishes the MIT-licensed Pro-RL checkpoint for local deployment.","bestUseCase":"Complex software engineering, long-running agents and multimodal professional work that needs more than Flash, with released weights for self-hosting at trillion-parameter scale.","pricingSummary":{"value":"Xiaomi MiMo API; input 0.435 USD/1M; cached read 0.0036/1M; output 0.87/1M","status":"verified","sources":[{"label":"MiMo-V2.6-Pro","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-pro"},{"label":"OpenRouter MiMo V2.6 Pro endpoints","url":"https://openrouter.ai/api/v1/models/xiaomi/mimo-v2.6-pro/endpoints"}]},"relatedArtifacts":[],"normalTokensPerSecond":115.68,"normalTokensPerSecondMin":114.47,"normalTokensPerSecondMax":116.45},"moonshotai/kimi-k3.1-pro":{"id":"moonshotai/kimi-k3.1-pro","name":"Kimi K3.1 Pro","facts":{"family":{"value":"Kimi K3.1","status":"verified","sources":[{"label":"Kimi K3.1 Pro quickstart","url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart"},{"label":"Kimi K3.1 Pro","url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"}]},"developer":{"value":"Moonshot AI","status":"verified","sources":[{"label":"Kimi K3.1 Pro","url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"},{"label":"Kimi K3.1 Pro quickstart","url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"About Moonshot AI","url":"https://www.moonshot.cn/about"}]},"releasedAt":{"value":"2026-10-23","status":"reviewed","sources":[{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"Kimi K3.1 Pro","url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"}]},"license":{"value":"Not yet published (weights due 4 Nov 2026)","status":"verified","sources":[{"label":"Kimi K3.1 Pro","url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"},{"label":"Kimi K3.1 Pro quickstart","url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart"}],"url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"},"architecture":{"value":"Post-trained version of Kimi K3, a Mixture-of-Experts model","status":"reviewed","sources":[{"label":"Kimi K3.1 Pro","url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"},{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},"totalParameters":{"value":2800000000000,"status":"inferred","sources":[{"label":"Kimi K3.1 Pro","url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"},{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Kimi API model inference pricing","url":"https://platform.kimi.ai/docs/pricing/chat"},{"label":"OpenRouter Kimi K3.1 Pro endpoints","url":"https://openrouter.ai/api/v1/models/moonshotai/kimi-k3.1-pro/endpoints"}]},"maxOutputTokens":{"value":1048576,"status":"verified","sources":[{"label":"Kimi K3.1 Pro quickstart: output limit","url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart"},{"label":"OpenRouter model catalog maximum-output pattern","url":"https://openrouter.ai/api/v1/models"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"Kimi K3.1 Pro quickstart","url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart"},{"label":"OpenRouter Kimi K3.1 Pro endpoints","url":"https://openrouter.ai/api/v1/models/moonshotai/kimi-k3.1-pro/endpoints"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"OpenRouter Kimi K3.1 Pro endpoints","url":"https://openrouter.ai/api/v1/models/moonshotai/kimi-k3.1-pro/endpoints"},{"label":"Kimi K3.1 Pro quickstart","url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart"}]},"activeParameters":{"value":104000000000,"status":"inferred","sources":[{"label":"Kimi K3.1 Pro","url":"https://www.kimi.com/en/blog/kimi-k3-1-pro"},{"label":"Kimi-K3","url":"https://huggingface.co/moonshotai/Kimi-K3"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/moonshotai/kimi-k3.1-pro","providerModelId":"moonshotai/kimi-k3.1-pro","verifiedAt":"2026-10-23","sources":[{"label":"OpenRouter Kimi K3.1 Pro endpoints","url":"https://openrouter.ai/api/v1/models/moonshotai/kimi-k3.1-pro/endpoints"},{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"}]},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"moonshotai/kimi-k3.1-pro","verifiedAt":"2026-10-23","sources":[{"label":"OpenRouter Kimi K3.1 Pro endpoints","url":"https://openrouter.ai/api/v1/models/moonshotai/kimi-k3.1-pro/endpoints"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/kimi-k3.1-pro","providerModelId":"moonshotai/kimi-k3.1-pro"},{"kind":"gateway","label":"Vercel AI Gateway Highspeed","url":"https://vercel.com/ai-gateway/models/kimi-k3.1-pro-highspeed","providerModelId":"moonshotai/kimi-k3.1-pro-highspeed"},{"kind":"api","label":"Kimi API","url":"https://platform.kimi.ai/docs/pricing/chat","providerModelId":"kimi-k3.1-pro","verifiedAt":"2026-10-23","sources":[{"label":"Kimi API model inference pricing","url":"https://platform.kimi.ai/docs/pricing/chat"},{"label":"Kimi K3.1 Pro quickstart","url":"https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"moonshotai/kimi-k3.1-pro","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.19999999999999998,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Kimi API (Moonshot AI)","providerModelId":"kimi-k3.1-pro","observedAt":"2026-10-23","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.2,"cachedWrite":2,"availability":"active","status":"verified","sources":[{"label":"Kimi API model inference pricing","url":"https://platform.kimi.ai/docs/pricing/chat"}]},{"provider":"Kimi API (Moonshot AI)","providerModelId":"kimi-k3.1-pro-highspeed","observedAt":"2026-10-23","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.2,"cachedWrite":4,"availability":"active","status":"verified","sources":[{"label":"Kimi API model inference pricing","url":"https://platform.kimi.ai/docs/pricing/chat"}]}],"speed":[],"related":[{"id":"moonshotai/kimi-k3","name":"Kimi K3","benchmarkUrl":"/benchmarks/models/kimi-k3"}],"deployment":{"weightFootprints":[],"componentStorage":{}},"special":"First model of Moonshot AI's K3.1 series: Kimi K3 with much larger post-training compute, a million-token context, image and video input and a faster Highspeed variant. Moonshot plans to release the weights by 4 November 2026.","bestUseCase":"Long-horizon coding agents and game builds, scripted 3D and CAD modeling in tools such as Blender, and financial document research. Highspeed suits interactive work at double the token price.","pricingSummary":{"value":"Kimi API (Moonshot AI); input 2 USD/1M; cached read 0.20/1M; output 10/1M. Highspeed: input 4, cached read 0.20, output 20","status":"verified","sources":[{"label":"Kimi API model inference pricing","url":"https://platform.kimi.ai/docs/pricing/chat"},{"label":"OpenRouter Kimi K3.1 Pro endpoints","url":"https://openrouter.ai/api/v1/models/moonshotai/kimi-k3.1-pro/endpoints"}]},"relatedArtifacts":[],"normalTokensPerSecond":116.93,"normalTokensPerSecondMin":110.97,"normalTokensPerSecondMax":120.82},"google/gemini-3.8-flash":{"id":"google/gemini-3.8-flash","name":"Gemini 3.8 Flash","facts":{"family":{"value":"Gemini 3.8","status":"verified","sources":[{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"},{"label":"Gemini 3.8 Flash launch","url":"https://blog.google/technology/google-deepmind/gemini-3-8-flash/"}]},"developer":{"value":"Google","status":"verified","sources":[{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"},{"label":"Gemini 3.8 Flash launch","url":"https://blog.google/technology/google-deepmind/gemini-3-8-flash/"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Google Terms of Service","url":"https://www.about.google/policies/terms/"}]},"releasedAt":{"value":"2026-09-02","status":"verified","sources":[{"label":"Gemini 3.8 Flash launch","url":"https://blog.google/technology/google-deepmind/gemini-3-8-flash/"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Gemini API Additional Terms of Service","url":"https://ai.google.dev/gemini-api/terms?hl=en"}],"url":"https://ai.google.dev/gemini-api/terms"},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"}]},"maxOutputTokens":{"value":65536,"status":"verified","sources":[{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"}]},"inputModalities":{"value":["text","image","video","audio","PDF"],"status":"verified","sources":[{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/google/gemini-3.8-flash","providerModelId":"google/gemini-3.8-flash"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"google/gemini-3.8-flash"},{"kind":"api","label":"Gemini Developer API","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash","providerModelId":"gemini-3.8-flash","verifiedAt":"2026-10-06","sources":[{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"google/gemini-3.8-flash","currency":"USD","unit":"per-1M-tokens","input":0.8250000000000001,"output":4.125,"cachedRead":0.0825,"cachedWrite":0.0416666666666667,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Gemini Developer API","providerModelId":"gemini-3.8-flash","observedAt":"2026-10-06","effectiveAt":"2026-09-02","currency":"USD","unit":"per-1M-tokens","input":0.75,"output":3.75,"cachedRead":0.075,"batchInput":0.375,"batchOutput":1.875,"availability":"active","status":"verified","sources":[{"label":"Gemini Developer API pricing","url":"https://ai.google.dev/gemini-api/docs/pricing"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"high","decodeTokensPerSecond":239.1,"timeToFirstTokenMs":15820,"reasoningModeOrBudget":"high","status":"verified","sources":[{"label":"Gemini 3.8 Flash (high) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gemini-3-8-flash"}]}],"related":[],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Gemini 3.8 Flash (high) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gemini-3-8-flash"},{"label":"Gemini 3.8 Flash","url":"https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"},{"label":"Gemini 3.8 Flash launch","url":"https://blog.google/technology/google-deepmind/gemini-3-8-flash/"}]},"weightFootprints":[],"componentStorage":{}},"special":"A fast hosted model that accepts text, images, audio and video, with a million-token-class context.","bestUseCase":"High-volume media analysis and agents that need to work across documents, recordings and video through one API.","pricingSummary":{"value":"Gemini Developer API; input 0.75 USD/1M; cached read 0.075/1M; output 3.75/1M; Introductory paid rate through 2026-12-31; published 2027 rates are higher. Free tier is also documented.","status":"verified","sources":[{"label":"Gemini Developer API pricing","url":"https://ai.google.dev/gemini-api/docs/pricing"}]},"speedSummary":{"value":"Artificial Analysis: 239.1 output tokens/s (high), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Gemini 3.8 Flash (high) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gemini-3-8-flash"}]},"relatedArtifacts":[],"normalTokensPerSecond":170.22,"normalTokensPerSecondMin":40.52,"normalTokensPerSecondMax":235.18},"openai/gpt-5.6-sol":{"id":"openai/gpt-5.6-sol","name":"GPT-5.6 Sol","facts":{"family":{"value":"GPT-5.6","status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-07-09","status":"verified","sources":[{"label":"GPT-5.6: Frontier intelligence that scales with your ambition","url":"https://openai.com/index/gpt-5-6/"},{"label":"OpenAI API changelog","url":"https://developers.openai.com/api/docs/changelog"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-5.6-sol","providerModelId":"openai/gpt-5.6-sol"},{"kind":"api","label":"OpenAI API","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol","providerModelId":"gpt-5.6-sol","verifiedAt":"2026-10-06","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-5.6-sol","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":4,"output":20,"cachedRead":0.4,"cachedWrite":5,"availability":"active","status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","decodeTokensPerSecond":96.6,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"GPT-5.6 Sol - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-6-sol"}]}],"related":[{"id":"openai/gpt-5.6-terra","name":"GPT-5.6 Terra","benchmarkUrl":"/benchmarks/models/gpt-5.6-terra"},{"id":"openai/gpt-5.6-luna","name":"GPT-5.6 Luna","benchmarkUrl":"/benchmarks/models/gpt-5.6-luna"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"},{"label":"All models","url":"https://developers.openai.com/api/docs/models/all"},{"label":"Deprecations","url":"https://developers.openai.com/api/docs/deprecations"},{"label":"GPT-5.6 Sol - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-6-sol"}]},"weightFootprints":[],"componentStorage":{}},"special":"The GPT-5.6 generation's high-capability Sol tier remains available, with a million-token-class context for coding and professional work.","bestUseCase":"Existing GPT-5.6 deployments that need Sol-level reasoning, or comparisons against newer GPT-6 models using the established model route.","pricingSummary":{"value":"OpenAI API; input 4 USD/1M; cached read 0.4/1M; output 20/1M; Promotional standard rates shown in first-party model page; >272K input triggers long-context multipliers.","status":"verified","sources":[{"label":"GPT-5.6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-sol"}]},"speedSummary":{"value":"Artificial Analysis: 96.6 output tokens/s (max), observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-5.6 Sol - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-6-sol"}]},"relatedArtifacts":[],"normalTokensPerSecond":81.37,"normalTokensPerSecondMin":56.77,"normalTokensPerSecondMax":130.33},"openai/gpt-5.6-terra":{"id":"openai/gpt-5.6-terra","name":"GPT-5.6 Terra","facts":{"family":{"value":"GPT-5.6","status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-07-09","status":"verified","sources":[{"label":"GPT-5.6: Frontier intelligence that scales with your ambition","url":"https://openai.com/index/gpt-5-6/"},{"label":"OpenAI API changelog","url":"https://developers.openai.com/api/docs/changelog"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-5.6-terra","providerModelId":"openai/gpt-5.6-terra"},{"kind":"api","label":"OpenAI API","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra","providerModelId":"gpt-5.6-terra","verifiedAt":"2026-10-06","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-5.6-terra","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":2,"output":12,"cachedRead":0.2,"cachedWrite":2.5,"availability":"active","status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"max","decodeTokensPerSecond":131.7,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"GPT-5.6 Terra - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-6-terra"}]}],"related":[{"id":"openai/gpt-5.6-sol","name":"GPT-5.6 Sol","benchmarkUrl":"/benchmarks/models/gpt-5.6-sol"},{"id":"openai/gpt-5.6-luna","name":"GPT-5.6 Luna","benchmarkUrl":"/benchmarks/models/gpt-5.6-luna"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"All models","url":"https://developers.openai.com/api/docs/models/all"},{"label":"Deprecations","url":"https://developers.openai.com/api/docs/deprecations"},{"label":"GPT-5.6 Terra - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-6-terra"},{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]},"weightFootprints":[],"componentStorage":{}},"special":"The middle GPT-5.6 tier balances capability and token price between Sol and Luna while retaining a million-token-class context.","bestUseCase":"Large-context production tasks that need more reasoning capability than Luna at a lower token cost than Sol.","pricingSummary":{"value":"OpenAI API; input 2 USD/1M; cached read 0.2/1M; output 12/1M","status":"verified","sources":[{"label":"GPT-5.6 Terra Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-terra"}]},"speedSummary":{"value":"Artificial Analysis: 131.7 output tokens/s (max), observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-5.6 Terra - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-6-terra"}]},"relatedArtifacts":[],"normalTokensPerSecond":101.5,"normalTokensPerSecondMin":43.03,"normalTokensPerSecondMax":195.24},"openai/gpt-6-luna":{"id":"openai/gpt-6-luna","name":"GPT-6 Luna","facts":{"family":{"value":"GPT-6","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-09-22","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-6-luna","providerModelId":"openai/gpt-6-luna"},{"kind":"api","label":"OpenAI API","url":"https://developers.openai.com/api/docs/models/gpt-6-luna","providerModelId":"gpt-6-luna","verifiedAt":"2026-10-06","sources":[{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-6-luna","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.1,"output":0.5,"cachedRead":0.01,"cachedWrite":0.125,"availability":"active","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"high","decodeTokensPerSecond":133.3,"reasoningModeOrBudget":"high","status":"verified","sources":[{"label":"GPT-6 Luna (high) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-6-luna-high"}]}],"related":[{"id":"openai/gpt-6.1-sol","name":"GPT-6.1 Sol","benchmarkUrl":"/benchmarks/models/gpt-6.1-sol"},{"id":"openai/gpt-6-astra","name":"GPT-6 Astra","benchmarkUrl":"/benchmarks/models/gpt-6-astra"},{"id":"openai/gpt-6-sol","name":"GPT-6 Sol","benchmarkUrl":"/benchmarks/models/gpt-6-sol"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"GPT-6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-6-luna"},{"label":"GPT-6 Luna (high) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-6-luna-high"}]},"weightFootprints":[],"componentStorage":{}},"special":"GPT-6's economical tier emphasizes fast output and low token cost while retaining a million-token-class context.","bestUseCase":"High-volume extraction, classification and agent substeps, with harder cases passed to a stronger model.","pricingSummary":{"value":"OpenAI API; input 0.1 USD/1M; cached read 0.01/1M; output 0.5/1M","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]},"speedSummary":{"value":"Artificial Analysis: 133.3 output tokens/s (high), observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-6 Luna (high) - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-6-luna-high"}]},"relatedArtifacts":[],"normalTokensPerSecond":131.4,"normalTokensPerSecondMin":56.12,"normalTokensPerSecondMax":174.34},"openai/gpt-6-sol":{"id":"openai/gpt-6-sol","name":"GPT-6 Sol","facts":{"family":{"value":"GPT-6","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-09-22","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-6-sol","providerModelId":"openai/gpt-6-sol"},{"kind":"api","label":"OpenAI API","url":"https://developers.openai.com/api/docs/models/gpt-6-sol","providerModelId":"gpt-6-sol","verifiedAt":"2026-10-06","sources":[{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-6-sol","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.2,"cachedWrite":2.5,"availability":"active","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"max","decodeTokensPerSecond":110,"reasoningModeOrBudget":"max","status":"verified","sources":[{"label":"GPT-6 Sol: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-sol"}]}],"related":[{"id":"openai/gpt-6.1-sol","name":"GPT-6.1 Sol","benchmarkUrl":"/benchmarks/models/gpt-6.1-sol"},{"id":"openai/gpt-6-astra","name":"GPT-6 Astra","benchmarkUrl":"/benchmarks/models/gpt-6-astra"},{"id":"openai/gpt-6-luna","name":"GPT-6 Luna","benchmarkUrl":"/benchmarks/models/gpt-6-luna"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"GPT-6 Sol Model","url":"https://developers.openai.com/api/docs/models/gpt-6-sol"},{"label":"All models","url":"https://developers.openai.com/api/docs/models/all"},{"label":"Deprecations","url":"https://developers.openai.com/api/docs/deprecations"},{"label":"GPT-6 Sol: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-sol"}]},"weightFootprints":[],"componentStorage":{}},"special":"The original GPT-6 Sol reasoning tier remains available for coding and agents alongside the newer GPT-6.1 Sol.","bestUseCase":"Maintaining established GPT-6 agent workflows and evaluating an upgrade to GPT-6.1 against a stable baseline.","pricingSummary":{"value":"OpenAI API; input 2 USD/1M; cached read 0.2/1M; output 10/1M","status":"verified","sources":[{"label":"Introducing GPT-6 Sol and Luna","url":"https://openai.com/index/introducing-gpt-6-sol-and-luna/"},{"label":"OpenAI API Pricing","url":"https://developers.openai.com/api/docs/pricing"}]},"speedSummary":{"value":"Artificial Analysis: 110 output tokens/s (max), observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-6 Sol: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-6-sol"}]},"relatedArtifacts":[],"normalTokensPerSecond":90.55,"normalTokensPerSecondMin":56.23,"normalTokensPerSecondMax":116.55},"x-ai/grok-4.7":{"id":"x-ai/grok-4.7","name":"Grok 4.7","facts":{"family":{"value":"Grok 4.7","status":"verified","sources":[{"label":"Introducing Grok 4.7","url":"https://x.ai/news/grok-4-7"},{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]},"developer":{"value":"SpaceXAI","status":"verified","sources":[{"label":"Introducing Grok 4.7","url":"https://x.ai/news/grok-4-7"},{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Terms of Service - Consumer","url":"https://x.ai/legal/terms-of-service"}]},"releasedAt":{"value":"2026-09-21","status":"verified","sources":[{"label":"Introducing Grok 4.7","url":"https://x.ai/news/grok-4-7"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Terms of Service - Enterprise","url":"https://x.ai/legal/terms-of-service-enterprise"}],"url":"https://x.ai/legal/terms-of-service-enterprise"},"contextWindowTokens":{"value":500000,"status":"verified","sources":[{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/x-ai/grok-4.7","providerModelId":"x-ai/grok-4.7"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"x-ai/grok-4.7"},{"kind":"api","label":"Grok API","url":"https://docs.x.ai/developers/models/grok-4.7","providerModelId":"grok-4.7","verifiedAt":"2026-10-06","sources":[{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]}],"pricing":[{"provider":"Grok API","providerModelId":"grok-4.7","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":2,"output":6,"cachedRead":0.5,"availability":"active","status":"verified","sources":[{"label":"Introducing Grok 4.7","url":"https://x.ai/news/grok-4-7"},{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]}],"speed":[{"provider":"Grok API","observedAt":"2026-10-06","status":"verified","sources":[{"label":"Introducing Grok 4.7","url":"https://x.ai/news/grok-4-7"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"xhigh","decodeTokensPerSecond":92.1,"timeToFirstTokenMs":43170,"reasoningModeOrBudget":"xhigh","status":"verified","sources":[{"label":"Grok 4.7 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/grok-4-7"}]}],"related":[],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Grok 4.7 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/grok-4-7"},{"label":"Introducing Grok 4.7","url":"https://x.ai/news/grok-4-7"},{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]},"weightFootprints":[],"componentStorage":{}},"special":"Hosted reasoning with a 500k-token context and competitive token rates. High reasoning effort can add substantial time before the first answer token.","bestUseCase":"Coding and knowledge-intensive analysis in xAI's ecosystem when deliberate reasoning matters more than an immediate first response.","pricingSummary":{"value":"Grok API; input 2 USD/1M; cached read 0.5/1M; output 6/1M","status":"verified","sources":[{"label":"Introducing Grok 4.7","url":"https://x.ai/news/grok-4-7"},{"label":"Grok 4.7","url":"https://docs.x.ai/developers/models/grok-4.7"}]},"speedSummary":{"value":"Artificial Analysis: 92.1 output tokens/s (xhigh), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Grok 4.7 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/grok-4-7"}]},"relatedArtifacts":[],"normalTokensPerSecond":185.38,"normalTokensPerSecondMin":173.34,"normalTokensPerSecondMax":196.9},"anthropic/claude-sonnet-5.5":{"id":"anthropic/claude-sonnet-5.5","name":"Claude Sonnet 5.5","facts":{"family":{"value":"Claude 5.5","status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"developer":{"value":"Anthropic","status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Anthropic Transparency Hub","url":"https://www.anthropic.com/transparency/voluntary-commitments"}]},"releasedAt":{"value":"2026-09-28","status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Anthropic Commercial Terms of Service","url":"https://www.anthropic.com/legal/commercial-terms"}],"url":"https://www.anthropic.com/legal/commercial-terms"},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/anthropic/claude-sonnet-5.5","providerModelId":"anthropic/claude-sonnet-5.5"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"anthropic/claude-sonnet-5.5"},{"kind":"api","label":"Anthropic API","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview","providerModelId":"claude-sonnet-5-5","verifiedAt":"2026-10-06","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"anthropic/claude-sonnet-5.5","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.09999999999999999,"cachedWrite":2.5,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Anthropic API","providerModelId":"claude-sonnet-5-5","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":2,"output":10,"cachedRead":0.2,"cachedWrite":2.5,"batchInput":1,"batchOutput":5,"availability":"active","status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]}],"speed":[{"provider":"Anthropic","observedAt":"2026-10-06","status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":127.5,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Claude Sonnet 5.5 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/claude-sonnet-5-5"}]}],"related":[{"id":"anthropic/claude-opus-5.5","name":"Claude Opus 5.5","benchmarkUrl":"/benchmarks/models/claude-opus-5.5"},{"id":"anthropic/claude-haiku-5.5","name":"Claude Haiku 5.5","benchmarkUrl":"/benchmarks/models/claude-haiku-5.5"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Claude Sonnet 5.5 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/claude-sonnet-5-5"},{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"weightFootprints":[],"componentStorage":{}},"special":"Pairs high-end Claude reasoning with faster measured output and lower token rates than Opus 5.5 and Fable 5.1.","bestUseCase":"Everyday production coding, document analysis and agents that need strong reasoning at a lower cost than Claude's premium tiers.","pricingSummary":{"value":"Anthropic API; input 2 USD/1M; cached read 0.2/1M; output 10/1M; 1-hour cache write $4/M; batch 50% discount.","status":"verified","sources":[{"label":"Claude Sonnet 5.5","url":"https://platform.claude.com/docs/en/models/sonnet-5-5/overview"}]},"speedSummary":{"value":"Artificial Analysis: 127.5 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Claude Sonnet 5.5 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/claude-sonnet-5-5"}]},"relatedArtifacts":[],"normalTokensPerSecond":248.74,"normalTokensPerSecondMin":241.45,"normalTokensPerSecondMax":260.99},"xiaomi/mimo-v2.6-flash":{"id":"xiaomi/mimo-v2.6-flash","name":"MiMo-V2.6-Flash","facts":{"family":{"value":"MiMo V2.6","status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"},{"label":"MiMo V2.6 release and open-source notice","url":"https://mimo.mi.com/docs/en-US/news/latest/v2-6"}]},"developer":{"value":"Xiaomi","status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"},{"label":"MiMo V2.6 release and open-source notice","url":"https://mimo.mi.com/docs/en-US/news/latest/v2-6"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"Xiaomi Launch May 2026 Terms and Conditions","url":"https://www.mi.com/global/support/terms/xiaomi-launch-may-2026/"}]},"releasedAt":{"value":"2026-09-21","status":"reviewed","sources":[{"label":"MiMo-V2.6-Flash-RL repository metadata","url":"https://huggingface.co/api/models/XiaomiMiMo/MiMo-V2.6-Flash-RL"},{"label":"Xiaomi MiMo","url":"https://en.wikipedia.org/wiki/Xiaomi_MiMo"}]},"license":{"value":"MIT","status":"reviewed","sources":[{"label":"Official MiMo V2.6 Flash model card","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"},{"label":"Xiaomi MiMo models","url":"https://huggingface.co/XiaomiMiMo/models"}],"url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"},"architecture":{"value":"Released Flash-RL checkpoint: sparse Mixture-of-Experts; hosted Flash and MOPD variants are distinct artifacts","status":"reviewed","sources":[{"label":"MiMo-V2.6-Flash-RL","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"},{"label":"MiMo-V2.6-Flash - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/mimo-v2-6-flash"}]},"totalParameters":{"value":309000000000,"status":"reviewed","sources":[{"label":"MiMo-V2.6-Flash-RL","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"},{"label":"MiMo-V2.6-Flash - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/mimo-v2-6-flash"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"}]},"inputModalities":{"value":["text","image","video","audio"],"status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"}]},"weightsAvailability":{"value":"official-open-variants","status":"reviewed","sources":[{"label":"Official MiMo V2.6 Flash model card","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"},{"label":"Xiaomi MiMo models","url":"https://huggingface.co/XiaomiMiMo/models"}]},"activeParameters":{"value":15000000000,"status":"reviewed","sources":[{"label":"MiMo-V2.6-Flash-RL","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"},{"label":"MiMo-V2.6-Flash - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/mimo-v2-6-flash"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/xiaomi/mimo-v2.6-flash","providerModelId":"xiaomi/mimo-v2.6-flash"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"xiaomi/mimo-v2.6-flash"},{"kind":"api","label":"Xiaomi MiMo API","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash","providerModelId":"mimo-v2.6-flash","verifiedAt":"2026-10-06","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"}]},{"kind":"download","label":"Official MiMo V2.6 Flash RL weights","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL","providerModelId":"XiaomiMiMo/MiMo-V2.6-Flash-RL","verifiedAt":"2026-10-06","sources":[{"label":"Xiaomi MiMo models","url":"https://huggingface.co/XiaomiMiMo/models"}]},{"kind":"download","label":"Official MiMo V2.6 Flash MOPD weights","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-MOPD","providerModelId":"XiaomiMiMo/MiMo-V2.6-Flash-MOPD","verifiedAt":"2026-10-06","sources":[{"label":"Xiaomi MiMo models","url":"https://huggingface.co/XiaomiMiMo/models"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"xiaomi/mimo-v2.6-flash","currency":"USD","unit":"per-1M-tokens","input":0.14,"output":0.28,"cachedRead":0.0028,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Xiaomi MiMo API","providerModelId":"mimo-v2.6-flash","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.14,"output":0.28,"cachedRead":0.0028,"availability":"active","status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":62,"timeToFirstTokenMs":4420,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"MiMo-V2.6-Flash - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/mimo-v2-6-flash"}]}],"related":[{"id":"xiaomi/mimo-v2.6-pro","name":"MiMo-V2.6-Pro","benchmarkUrl":"/benchmarks/models/mimo-v2.6-pro"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Xiaomi MiMo models","url":"https://huggingface.co/XiaomiMiMo/models"},{"label":"MiMo-V2.6-Flash-RL","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"}]},"deploymentClass":{"value":"multi-GPU datacenter","status":"inferred","sources":[{"label":"Xiaomi MiMo models","url":"https://huggingface.co/XiaomiMiMo/models"},{"label":"MiMo-V2.6-Flash-RL","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"}]},"weightFootprints":[{"artifact":"Parameter-based FP8 weight-only lower bound","precisionOrQuantization":"FP8 equivalent","sizeGB":309,"status":"inferred","sources":[{"label":"MiMo-V2.6-Flash-RL","url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL"}]}],"componentStorage":{}},"special":"Low-cost multimodal Flash service with a million-token context. Xiaomi also publishes MIT-licensed RL and MOPD checkpoints for local deployment.","bestUseCase":"Multimodal agents and high-volume API work where token cost matters, with a separate choice of released checkpoint for self-hosting.","pricingSummary":{"value":"Xiaomi MiMo API; input 0.14 USD/1M; cached read 0.0028/1M; output 0.28/1M","status":"verified","sources":[{"label":"MiMo-V2.6-Flash","url":"https://mimo.mi.com/models/en-US/mimo-v2.6-flash"}]},"speedSummary":{"value":"Artificial Analysis: 62 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"MiMo-V2.6-Flash - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/mimo-v2-6-flash"}]},"relatedArtifacts":[],"normalTokensPerSecond":547.11,"normalTokensPerSecondMin":387.95,"normalTokensPerSecondMax":632.3},"qwen/qwen3.8-27B":{"id":"qwen/qwen3.8-27B","name":"Qwen3.8-27B","facts":{"family":{"value":"Qwen 3.8","status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"},{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]},"developer":{"value":"Qwen","status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"},{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]},"origin":{"value":"China (Alibaba Group/Qwen organization context)","status":"verified","sources":[{"label":"Alibaba Group corporate information FAQ","url":"https://www.alibabagroup.com/en-US/faqs-corporate-information"}]},"releasedAt":{"value":"2026-08-14","status":"verified","sources":[{"label":"Qwen3.8 repository news","url":"https://github.com/QwenLM/Qwen3.8"}]},"license":{"value":"Apache-2.0","status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"}],"url":"https://huggingface.co/Qwen/Qwen3.8-27B"},"architecture":{"value":"Dense causal language model with vision encoder; hybrid Gated DeltaNet and gated attention; MTP-trained","status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"}]},"totalParameters":{"value":27000000000,"status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"},{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]},"maxOutputTokens":{"value":131072,"status":"verified","sources":[{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"},{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"},{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"}]},"activeParameters":{"value":27000000000,"status":"inferred","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/qwen/qwen3.8-27B","providerModelId":"qwen/qwen3.8-27B"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"qwen/qwen3.8-27B"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/Qwen/Qwen3.8-27B","providerModelId":"Qwen/Qwen3.8-27B","verifiedAt":"2026-10-06","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"}]},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF","providerModelId":"unsloth/Qwen3.8-27B-GGUF","verifiedAt":"2026-10-06","sources":[{"label":"Qwen3.8-27B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF"}]},{"kind":"api","label":"Alibaba Cloud Model Studio","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b","providerModelId":"qwen3.8-27b","verifiedAt":"2026-10-06","sources":[{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]}],"pricing":[{"provider":"Alibaba Cloud Model Studio","providerModelId":"qwen3.8-27b","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.424,"output":1.696,"cachedRead":0.085,"cachedWrite":0.53,"availability":"active","status":"verified","sources":[{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":45.8,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"Qwen3.8 27B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-8-27b"}]}],"related":[{"id":"qwen/qwen3.8-flash","name":"Qwen3.8-Flash","benchmarkUrl":"/benchmarks/models/qwen3.8-flash"},{"id":"qwen/qwen3.8-max-0902","name":"Qwen3.8-Max-0902","benchmarkUrl":"/benchmarks/models/qwen3.8-max-0902"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"},{"label":"Qwen3.8-27B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF"}]},"deploymentClass":{"value":"single high-memory workstation GPU","status":"inferred","sources":[{"label":"Qwen3.8-27B","url":"https://huggingface.co/Qwen/Qwen3.8-27B"},{"label":"Qwen3.8-27B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF"}]},"weightFootprints":[{"artifact":"Unsloth Q8_0","precisionOrQuantization":"8-bit","sizeGB":29,"status":"verified","sources":[{"label":"Qwen3.8-27B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF"}]},{"artifact":"Unsloth BF16","precisionOrQuantization":"BF16","sizeGB":54.7,"status":"verified","sources":[{"label":"Qwen3.8-27B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF"}]},{"artifact":"MTP Q4_0 component","precisionOrQuantization":"4-bit","sizeGB":1.37,"status":"verified","sources":[{"label":"Qwen3.8-27B-GGUF","url":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF"}]}],"componentStorage":{}},"special":"Apache-2.0 multimodal model with 27B parameters and Q4 weight files around 14–18 GB, within reach of a high-memory workstation.","bestUseCase":"Local coding assistants and multimodal agents where control over the weights matters and a smaller model leaves too much capability behind.","pricingSummary":{"value":"Alibaba Cloud Model Studio; input 0.424 USD/1M; cached read 0.085/1M; output 1.696/1M; Representative documented regional rate; cache and region-specific rates are separate.","status":"verified","sources":[{"label":"qwen3.8-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-8-27b"}]},"speedSummary":{"value":"Artificial Analysis: 45.8 output tokens/s (current), observed 2026-10-06.","status":"reviewed","sources":[{"label":"Qwen3.8 27B - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/qwen3-8-27b"}]},"relatedArtifacts":[],"normalTokensPerSecond":78.45,"normalTokensPerSecondMin":74.08,"normalTokensPerSecondMax":81.07},"openai/gpt-5.5":{"id":"openai/gpt-5.5","name":"GPT-5.5","facts":{"family":{"value":"GPT-5.5","status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-04-23","status":"verified","sources":[{"label":"Introducing GPT-5.5","url":"https://openai.com/index/introducing-gpt-5-5/"},{"label":"OpenAI API changelog","url":"https://developers.openai.com/api/docs/changelog"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-5.5","providerModelId":"openai/gpt-5.5"},{"kind":"api","label":"OpenAI API","url":"https://developers.openai.com/api/docs/models/gpt-5.5","providerModelId":"gpt-5.5","verifiedAt":"2026-10-06","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-5.5","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":5,"output":30,"cachedRead":0.5,"availability":"active","status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","decodeTokensPerSecond":104.2,"timeToFirstTokenMs":43820,"status":"verified","sources":[{"label":"GPT-5.5 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-5"}]}],"related":[],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"},{"label":"All models","url":"https://developers.openai.com/api/docs/models/all"},{"label":"Deprecations","url":"https://developers.openai.com/api/docs/deprecations"},{"label":"GPT-5.5 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-5"}]},"weightFootprints":[],"componentStorage":{}},"special":"An earlier OpenAI flagship generation that remains available for coding and professional work, with a million-token-class context.","bestUseCase":"Maintaining applications already evaluated on GPT-5.5 and measuring whether newer models improve their results.","pricingSummary":{"value":"OpenAI API; input 5 USD/1M; cached read 0.5/1M; output 30/1M; Model page lists alias gpt-5.5 and snapshot gpt-5.5-2026-04-23.","status":"verified","sources":[{"label":"GPT-5.5 Model","url":"https://developers.openai.com/api/docs/models/gpt-5.5"}]},"speedSummary":{"value":"Artificial Analysis: 104.2 output tokens/s, observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-5.5 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/gpt-5-5"}]},"relatedArtifacts":[],"normalTokensPerSecond":192.97,"normalTokensPerSecondMin":189.1,"normalTokensPerSecondMax":195.7},"openai/gpt-5.6-luna":{"id":"openai/gpt-5.6-luna","name":"GPT-5.6 Luna","facts":{"family":{"value":"GPT-5.6","status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]},"developer":{"value":"OpenAI","status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"EU terms of use","url":"https://openai.com/policies/oct-2024-eu-terms/"}]},"releasedAt":{"value":"2026-07-09","status":"verified","sources":[{"label":"GPT-5.6: Frontier intelligence that scales with your ambition","url":"https://openai.com/index/gpt-5-6/"},{"label":"OpenAI API changelog","url":"https://developers.openai.com/api/docs/changelog"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"OpenAI Services Agreement","url":"https://openai.com/policies/services-agreement/"}],"url":"https://openai.com/policies/services-agreement/"},"contextWindowTokens":{"value":1050000,"status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]},"maxOutputTokens":{"value":128000,"status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/openai/gpt-5.6-luna","providerModelId":"openai/gpt-5.6-luna"},{"kind":"api","label":"OpenAI API","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna","providerModelId":"gpt-5.6-luna","verifiedAt":"2026-10-06","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]}],"pricing":[{"provider":"OpenAI API","providerModelId":"gpt-5.6-luna","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.2,"output":1.2,"cachedRead":0.02,"cachedWrite":0.25,"availability":"active","status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]}],"speed":[{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"release comparison","decodeTokensPerSecond":127.9,"reasoningModeOrBudget":"release comparison","status":"verified","sources":[{"label":"GPT-5.6 Luna: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-5-6-luna"}]}],"related":[{"id":"openai/gpt-5.6-sol","name":"GPT-5.6 Sol","benchmarkUrl":"/benchmarks/models/gpt-5.6-sol"},{"id":"openai/gpt-5.6-terra","name":"GPT-5.6 Terra","benchmarkUrl":"/benchmarks/models/gpt-5.6-terra"}],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"All models","url":"https://developers.openai.com/api/docs/models/all"},{"label":"Deprecations","url":"https://developers.openai.com/api/docs/deprecations"},{"label":"GPT-5.6 Luna: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-5-6-luna"},{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]},"weightFootprints":[],"componentStorage":{}},"special":"The economical GPT-5.6 tier combines fast measured output with low token rates and a million-token-class context.","bestUseCase":"High-volume extraction, classification and smaller agent steps in existing GPT-5.6 applications.","pricingSummary":{"value":"OpenAI API; input 0.2 USD/1M; cached read 0.02/1M; output 1.2/1M","status":"verified","sources":[{"label":"GPT-5.6 Luna Model","url":"https://developers.openai.com/api/docs/models/gpt-5.6-luna"}]},"speedSummary":{"value":"Artificial Analysis: 127.9 output tokens/s (release comparison), observed 2026-10-06.","status":"reviewed","sources":[{"label":"GPT-5.6 Luna: Release Intelligence, Performance & Price","url":"https://artificialanalysis.ai/models/releases/gpt-5-6-luna"}]},"relatedArtifacts":[],"normalTokensPerSecond":180.51,"normalTokensPerSecondMin":56.06,"normalTokensPerSecondMax":267.52},"mistralai/mistral-large-4-0":{"id":"mistralai/mistral-large-4-0","name":"Mistral Large 4","facts":{"family":{"value":"Mistral","status":"verified","sources":[{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"},{"label":"Introducing Mistral Large 4","url":"https://mistral.ai/news/mistral-large-4/"}]},"developer":{"value":"Mistral AI","status":"verified","sources":[{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"},{"label":"Introducing Mistral Large 4","url":"https://mistral.ai/news/mistral-large-4/"}]},"origin":{"value":"France","status":"verified","sources":[{"label":"Mistral AI Commercial Terms of Service","url":"https://legal.mistral.ai/terms/commercial-terms-of-service"}]},"releasedAt":{"value":"2026-10-06","status":"verified","sources":[{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"},{"label":"Introducing Mistral Large 4","url":"https://mistral.ai/news/mistral-large-4/"}]},"license":{"value":"Hosted service; weights license not established","status":"verified","sources":[{"label":"Introducing Mistral Large 4","url":"https://mistral.ai/news/mistral-large-4/"},{"label":"Mistral Large 4 upcoming release on Hugging Face","url":"https://huggingface.co/mistralai/Mistral-Large-4.0-1T05-A52B"},{"label":"Mistral AI Commercial Terms of Service","url":"https://legal.mistral.ai/terms/commercial-terms-of-service"}],"url":"https://legal.mistral.ai/terms/commercial-terms-of-service"},"architecture":{"value":"Granular mixture of experts (MoE); multimodal text and image input, text output.","status":"verified","sources":[{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"},{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"}]},"totalParameters":{"value":1050000000000,"status":"verified","sources":[{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"}]},"contextWindowTokens":{"value":1048576,"status":"verified","sources":[{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"},{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"}]},"maxOutputTokens":{"value":262144,"status":"verified","sources":[{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"}]},"activeParameters":{"value":49000000000,"status":"verified","sources":[{"label":"Introducing Mistral Large 4","url":"https://mistral.ai/news/mistral-large-4/"},{"label":"mistralai/Mistral-Large-4.0-1T05-A52B · Upcoming release · Hugging Face","url":"https://huggingface.co/mistralai/Mistral-Large-4.0-1T05-A52B"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/mistralai/mistral-large-4-0","providerModelId":"mistralai/mistral-large-4-0","verifiedAt":"2026-10-06","sources":[{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"}]},{"kind":"api","label":"Mistral Chat Completions","url":"https://api.mistral.ai/v1/chat/completions","providerModelId":"mistral-large-4","verifiedAt":"2026-10-06","sources":[{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"mistralai/mistral-large-4-0","currency":"USD","unit":"per-1M-tokens","input":1.3599999999999999,"output":4.18,"cachedRead":0.14,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"OpenRouter","providerModelId":"mistralai/mistral-large-4-0","observedAt":"2026-10-06T18:15:43.683Z","currency":"USD","unit":"per-1M-tokens","input":0.68,"output":2.09,"cachedRead":0.07,"availability":"Historical rates","status":"verified","sources":[{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"}]},{"provider":"Mistral AI API","providerModelId":"mistral-large-4","observedAt":"2026-10-23","currency":"USD","unit":"per-1M-tokens","input":1.36,"output":4.18,"cachedRead":0.14,"availability":"active","status":"verified","sources":[{"label":"Mistral Large 4 - Mistral AI | Mistral Docs","url":"https://docs.mistral.ai/models/mistral-large-4-0"},{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"}]}],"speed":[],"related":[],"deployment":{"weightFootprints":[],"componentStorage":{}},"special":"Multimodal MoE with text and image input, adjustable reasoning, tool calling and structured outputs. Available as a public API preview.","bestUseCase":"Coding agents and document analysis that combine image input, long context, tool calling and structured responses.","pricingSummary":{"value":"Recorded 2026-10-06: OpenRouter preview (2026-10-06): USD $0.68 input, $0.07 cached input read and $2.09 output per 1M tokens.","status":"verified","sources":[{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"},{"label":"OpenRouter Mistral Large 4 endpoints","url":"https://openrouter.ai/api/v1/models/mistralai/mistral-large-4-0/endpoints"}]},"relatedArtifacts":[],"normalTokensPerSecond":150.53,"normalTokensPerSecondMin":140.26,"normalTokensPerSecondMax":158.36},"qwen/qwen3.6-27b":{"id":"qwen/qwen3.6-27b","name":"Qwen3.6-27B","facts":{"family":{"value":"Qwen 3.6","status":"verified","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"}]},"developer":{"value":"Qwen / Alibaba","status":"verified","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"},{"label":"Alibaba Group corporate information FAQ","url":"https://www.alibabagroup.com/en-US/faqs-corporate-information"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"Alibaba Group corporate information FAQ","url":"https://www.alibabagroup.com/en-US/faqs-corporate-information"}]},"releasedAt":{"value":"2026-04-22","status":"reviewed","sources":[{"label":"Qwen3.6-27B Delivers Flagship Coding in 27B Dense Model","url":"https://letsdatascience.com/news/qwen36-27b-delivers-flagship-coding-in-27b-dense-model-fd698bba"},{"label":"OpenRouter public model registry","url":"https://openrouter.ai/api/v1/models"}]},"license":{"value":"Apache-2.0","status":"verified","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"},{"label":"Qwen3.6-27B checkpoint parameter metadata","url":"https://huggingface.co/api/models/Qwen/Qwen3.6-27B"}],"url":"https://huggingface.co/Qwen/Qwen3.6-27B/blob/main/LICENSE"},"architecture":{"value":"Dense causal language model with vision encoder; hybrid Gated DeltaNet and gated attention; MTP-trained","status":"verified","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"}]},"totalParameters":{"value":27000000000,"status":"verified","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"}]},"contextWindowTokens":{"value":1010000,"status":"verified","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"},{"label":"qwen3.6-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-27b"}]},"maxOutputTokens":{"value":65536,"status":"verified","sources":[{"label":"qwen3.6-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-27b"}]},"inputModalities":{"value":["text","image","video"],"status":"verified","sources":[{"label":"qwen3.6-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-27b"},{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"qwen3.6-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-27b"},{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"},{"label":"Qwen3.6-27B checkpoint parameter metadata","url":"https://huggingface.co/api/models/Qwen/Qwen3.6-27B"}]},"activeParameters":{"value":27000000000,"status":"reviewed","sources":[{"label":"Qwen3.6-27B model card","url":"https://huggingface.co/Qwen/Qwen3.6-27B"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/qwen/qwen3.6-27b","providerModelId":"qwen/qwen3.6-27b"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"qwen/qwen3.6-27b"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/Qwen/Qwen3.6-27B","providerModelId":"Qwen/Qwen3.6-27B"},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/Qwen3.6-27B-GGUF","providerModelId":"unsloth/Qwen3.6-27B-GGUF"}],"pricing":[{"provider":"OpenRouter","providerModelId":"qwen/qwen3.6-27b","currency":"USD","unit":"per-1M-tokens","input":0.25,"output":2,"cachedRead":0.024999999999999998,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Alibaba Cloud Model Studio","providerModelId":"qwen3.6-27b","observedAt":"2026-10-23","currency":"USD","unit":"per-1M-tokens","input":0.6,"output":3.6,"availability":"active","status":"verified","sources":[{"label":"qwen3.6-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-27b"}]},{"provider":"Alibaba Cloud Model Studio","providerModelId":"qwen3.6-27b","observedAt":"2026-10-23","currency":"USD","unit":"per-1M-tokens","input":0.412564,"output":2.475384,"availability":"active","status":"verified","sources":[{"label":"qwen3.6-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-27b"}]}],"speed":[],"related":[{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen3.6-35B-A3B","benchmarkUrl":"/benchmarks/models/qwen3.6-35b-a3b"}],"deployment":{"weightFootprints":[],"componentStorage":{}},"special":"An Apache-2.0 dense multimodal model with text, image and video input and reasoning-context preservation for iterative coding.","bestUseCase":"Local or hosted coding assistants, repository edits and multimodal agent workflows.","pricingSummary":{"value":"Alibaba Cloud Model Studio, international (Singapore): input 0.60 USD/1M; output 3.60/1M. China (Beijing): input 0.41; output 2.48","status":"verified","sources":[{"label":"qwen3.6-27b Model Info","url":"https://www.alibabacloud.com/help/en/model-studio/qwen3-6-27b"}]},"relatedArtifacts":[],"normalTokensPerSecond":95.65,"normalTokensPerSecondMin":91.96,"normalTokensPerSecondMax":98.22},"microsoft/mai-thinking-1":{"id":"microsoft/mai-thinking-1","name":"MAI-Thinking-1","facts":{"family":{"value":"MAI-Thinking","status":"verified","sources":[{"label":"Introducing MAI-Thinking-1","url":"https://microsoft.ai/news/introducing-mai-thinking-1/"}]},"developer":{"value":"Microsoft AI","status":"verified","sources":[{"label":"Introducing MAI-Thinking-1","url":"https://microsoft.ai/news/introducing-mai-thinking-1/"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Facts About Microsoft","url":"https://news.microsoft.com/facts-about-microsoft/"},{"label":"MAI-Thinking-1","url":"https://microsoft.ai/models/mai-thinking-1/"}]},"releasedAt":{"value":"2026-08-12","status":"verified","sources":[{"label":"Introducing MAI-Thinking-1","url":"https://microsoft.ai/news/introducing-mai-thinking-1/"}]},"license":{"value":"Proprietary hosted service","status":"reviewed","sources":[{"label":"Deploy and use MAI-Thinking-1 in Microsoft Foundry (preview)","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking"}]},"architecture":{"value":"Sparse Mixture-of-Experts (MoE)","status":"verified","sources":[{"label":"Introducing MAI-Thinking-1","url":"https://microsoft.ai/news/introducing-mai-thinking-1/"}]},"totalParameters":{"value":1000000000000,"status":"inferred","sources":[{"label":"MAI-Thinking-1","url":"https://microsoft.ai/models/mai-thinking-1/"},{"label":"Introducing MAI-Thinking-1","url":"https://microsoft.ai/news/introducing-mai-thinking-1/"}]},"contextWindowTokens":{"value":256000,"status":"inferred","sources":[{"label":"Deploy and use MAI-Thinking-1 in Microsoft Foundry (preview)","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking"}]},"maxOutputTokens":{"value":64000,"status":"inferred","sources":[{"label":"Deploy and use MAI-Thinking-1 in Microsoft Foundry (preview)","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Deploy and use MAI-Thinking-1 in Microsoft Foundry (preview)","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Deploy and use MAI-Thinking-1 in Microsoft Foundry (preview)","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking"}]},"activeParameters":{"value":35000000000,"status":"verified","sources":[{"label":"Introducing MAI-Thinking-1","url":"https://microsoft.ai/news/introducing-mai-thinking-1/"}]}},"access":[{"kind":"api","label":"Microsoft Foundry deployment and API guide","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking","providerModelId":"mai-thinking-1","verifiedAt":"2026-10-09","sources":[{"label":"Deploy and use MAI-Thinking-1 in Microsoft Foundry (preview)","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking"}]}],"pricing":[{"provider":"Microsoft Foundry","providerModelId":"mai-thinking-1","observedAt":"2026-10-23","effectiveAt":"2026-08-01","currency":"USD","unit":"per-1M-tokens","input":2,"output":8,"cachedRead":0.2,"availability":"public-preview","status":"verified","sources":[{"label":"Azure Retail Prices API: MAI-Thinking-1 meters","url":"https://prices.azure.com/api/retail/prices?api-version=2023-01-01-preview&$filter=contains(meterName,%20%27MAI-Thinking-1%27)"}]}],"speed":[],"related":[],"deployment":{"deploymentClass":{"value":"Hosted Microsoft Foundry deployment","status":"verified","sources":[{"label":"Deploy and use MAI-Thinking-1 in Microsoft Foundry (preview)","url":"https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/use-foundry-models-mai-thinking"}]},"weightFootprints":[],"componentStorage":{}},"special":"Microsoft AI reasoning model, with a sparse MoE architecture, 35B active parameters and function calling.","bestUseCase":"Multi-step math, coding agents and long-document enterprise workflows.","pricingSummary":{"value":"Microsoft Foundry, Global Standard; input 2 USD/1M; cached read 0.20/1M; output 8/1M","status":"verified","sources":[{"label":"Azure Retail Prices API: MAI-Thinking-1 meters","url":"https://prices.azure.com/api/retail/prices?api-version=2023-01-01-preview&$filter=contains(meterName,%20%27MAI-Thinking-1%27)"}]},"relatedArtifacts":[],"normalTokensPerSecond":93.41,"normalTokensPerSecondMin":86.89,"normalTokensPerSecondMax":105.63},"union-alpha":{"routerStatus":"confirmed","id":"union-alpha","name":"Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)","facts":{"family":{"value":"Pareto","status":"verified","sources":[{"label":"Pareto model card","url":"https://unbiased.ai/model-card/"},{"label":"How a Pareto answer gets made","url":"https://unbiased.ai/how/"},{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]},"developer":{"value":"Circuit & Chisel / Unbiased","status":"verified","sources":[{"label":"Pareto model card","url":"https://unbiased.ai/model-card/"},{"label":"How a Pareto answer gets made","url":"https://unbiased.ai/how/"},{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]},"origin":{"value":"United States","status":"verified","sources":[{"label":"Terms & Policies","url":"https://unbiased.ai/terms/"}]},"releasedAt":{"value":"2026-09-20","status":"verified","sources":[{"label":"Pareto model card","url":"https://unbiased.ai/model-card/"}]},"contextWindowTokens":{"value":262000,"status":"verified","sources":[{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]},"maxOutputTokens":{"value":131000,"status":"verified","sources":[{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"Pareto model card","url":"https://unbiased.ai/model-card/"},{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Pareto model card","url":"https://unbiased.ai/model-card/"},{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/unbiased/pareto","providerModelId":"unbiased/pareto"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"unbiased/pareto"}],"pricing":[{"provider":"Unbiased","providerModelId":"unbiased/pareto","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":2.5,"output":7.5,"cachedRead":0.25,"availability":"historical-version observation","status":"verified","sources":[{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]}],"speed":[{"provider":"AI BENCHY via Union Alpha guide","observedAt":"2026-10-06","modelSnapshot":"Pareto 26.9 / Union Alpha","endToEndLatencySeconds":31.51,"status":"verified","sources":[{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}],"conditions":"Average end-to-end response time across 22 tests."}],"related":[],"deployment":{"deploymentClass":{"value":"hosted-only","status":"verified","sources":[{"label":"Pareto 26.10 Preview","url":"https://unbiased.ai/blog/pareto-26-10-preview/"},{"label":"Changelog","url":"https://unbiased.ai/changelog/"}]},"weightFootprints":[],"componentStorage":{}},"special":"Pareto 26.9, first seen as Union Alpha, combines multiple models in one hosted system. This record retains that version's identity and benchmark history.","bestUseCase":"Comparing the 26.9 system's cost and capability with individual models. Check the selected API version because the unversioned route has advanced.","pricingSummary":{"value":"Pareto 26.9 historical rates: USD $2.50 input, $0.25 cached input and $7.50 output per 1M tokens. The unversioned route now serves 26.10 Preview.","status":"reviewed","sources":[{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"},{"label":"Pareto 26.10 Preview","url":"https://unbiased.ai/blog/pareto-26-10-preview/"},{"label":"Changelog","url":"https://unbiased.ai/changelog/"}]},"speedSummary":{"value":"Pareto 26.9: 31.51 s average end-to-end response time in the 22-test suite; this is latency, not decoding tokens/s.","status":"reviewed","sources":[{"label":"Union Alpha / Pareto identity guide","url":"https://union-alpha.com/"}]},"relatedArtifacts":[],"normalTokensPerSecond":25.74,"normalTokensPerSecondMin":21.62,"normalTokensPerSecondMax":29.47},"pixel-canary":{"id":"pixel-canary","name":"Callstack Apex","facts":{"family":{"value":"Apex","status":"verified","sources":[{"label":"Apex Is Now Generally Available: Agentic Coding Model for Mobile and Web, Specialized in React","url":"https://www.callstack.com/blog/apex-agetic-coding-for-mobile-and-web-spcialied-in-react"}]},"developer":{"value":"Callstack","status":"verified","sources":[{"label":"Apex Is Now Generally Available: Agentic Coding Model for Mobile and Web, Specialized in React","url":"https://www.callstack.com/blog/apex-agetic-coding-for-mobile-and-web-spcialied-in-react"}]},"origin":{"value":"Poland","status":"verified","sources":[{"label":"Apex Is Now Generally Available: Agentic Coding Model for Mobile and Web, Specialized in React","url":"https://www.callstack.com/blog/apex-agetic-coding-for-mobile-and-web-spcialied-in-react"}]},"releasedAt":{"value":"2026-10-01","status":"verified","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}]},"license":{"value":"Proprietary hosted service","status":"verified","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}],"url":"https://apex.callstack.com/models"},"architecture":{"value":"Qwen foundation with Callstack domain post-training for React and React Native","status":"verified","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}]},"contextWindowTokens":{"value":262144,"status":"verified","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}]},"weightsAvailability":{"value":"not-distributed","status":"verified","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}]}},"access":[{"kind":"api","label":"Callstack Apex API","url":"https://api.callstack.ai/","providerModelId":"apex-20261001","verifiedAt":"2026-10-13","sources":[{"label":"Apex Models","url":"https://apex.callstack.com/models"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/apex","providerModelId":"callstack/apex","verifiedAt":"2026-10-13","sources":[{"label":"Apex API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/apex"}]}],"pricing":[{"provider":"Vercel AI Gateway","providerModelId":"callstack/apex","currency":"USD","unit":"per-1M-tokens","input":0.5,"output":3,"cachedRead":0.19999999999999998,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"Vercel AI Gateway model pricing feed","url":"https://ai-gateway.vercel.sh/v1/models"}]},{"provider":"Callstack","providerModelId":"apex-20261001","observedAt":"2026-10-13","currency":"USD","unit":"per-1M-tokens","input":0.5,"output":3,"cachedRead":0.2,"availability":"active","status":"verified","sources":[{"label":"Apex Is Now Generally Available: Agentic Coding Model for Mobile and Web, Specialized in React","url":"https://www.callstack.com/blog/apex-agetic-coding-for-mobile-and-web-spcialied-in-react"}]}],"speed":[],"related":[],"deployment":{"deploymentClass":{"value":"Hosted API; commercial customer-hosted deployments arranged with Callstack","status":"verified","sources":[{"label":"Apex Is Now Generally Available: Agentic Coding Model for Mobile and Web, Specialized in React","url":"https://www.callstack.com/blog/apex-agetic-coding-for-mobile-and-web-spcialied-in-react"},{"label":"Apex Models","url":"https://apex.callstack.com/models"}]},"weightFootprints":[],"componentStorage":{}},"special":"Callstack’s Qwen-based coding model, revealed from Pixel Canary and specialized for React Native, React and Next.js.","bestUseCase":"Building and reviewing React Native apps, integrating native modules, and planning React migrations.","pricingSummary":{"value":"USD $0.50 input, $0.20 cached input and $3.00 output per 1M tokens.","status":"verified","sources":[{"label":"Apex Is Now Generally Available: Agentic Coding Model for Mobile and Web, Specialized in React","url":"https://www.callstack.com/blog/apex-agetic-coding-for-mobile-and-web-spcialied-in-react"},{"label":"Apex API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/apex"}]},"relatedArtifacts":[],"normalTokensPerSecond":255.19,"normalTokensPerSecondMin":252.71,"normalTokensPerSecondMax":259.5},"ox-alpha":{"id":"ox-alpha","name":"GLM-5.3-Flash","facts":{"family":{"value":"GLM-5","status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"},{"label":"GLM-5.3-Flash LICENSE","url":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE"}]},"developer":{"value":"Z.AI","status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"},{"label":"GLM-5.3-Flash LICENSE","url":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE"}]},"releasedAt":{"value":"2026-08-26","status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"license":{"value":"MIT","status":"verified","sources":[{"label":"GLM-5.3-Flash LICENSE","url":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE"}],"url":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE"},"architecture":{"value":"Hybrid sparse and linear attention MoE","status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"}]},"totalParameters":{"value":320000000000,"status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"maxOutputTokens":{"value":131000,"status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"inputModalities":{"value":["text","image"],"status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"GLM-5.3-Flash LICENSE","url":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE"}]},"activeParameters":{"value":18000000000,"status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/z-ai/glm-5.3-flash","providerModelId":"z-ai/glm-5.3-flash"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"z-ai/glm-5.3-flash"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/zai-org/GLM-5.3-Flash","providerModelId":"zai-org/GLM-5.3-Flash","verifiedAt":"2026-10-06","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"},{"label":"GLM-5.3-Flash LICENSE","url":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE"}]},{"kind":"gguf","label":"Unsloth GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF","providerModelId":"unsloth/GLM-5.3-Flash-GGUF","verifiedAt":"2026-10-06","sources":[{"label":"GLM-5.3-Flash-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF"}]},{"kind":"gateway","label":"Vercel AI Gateway","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash","providerModelId":"zai/glm-5.3-flash","verifiedAt":"2026-10-06","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"z-ai/glm-5.3-flash","currency":"USD","unit":"per-1M-tokens","input":0.15,"output":0.5,"cachedRead":0.03,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"Vercel AI Gateway / DigitalOcean route","providerModelId":"zai/glm-5.3-flash","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.15,"output":0.5,"cachedRead":0.03,"availability":"active","status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]}],"speed":[{"provider":"Vercel AI Gateway / DigitalOcean","observedAt":"2026-10-06","modelSnapshot":"zai/glm-5.3-flash","decodeTokensPerSecond":119,"timeToFirstTokenMs":500,"status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},{"provider":"Artificial Analysis","observedAt":"2026-10-06","modelSnapshot":"current","decodeTokensPerSecond":53,"reasoningModeOrBudget":"current","status":"verified","sources":[{"label":"GLM 5.3 Flash - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/glm-5-3-flash/"}]}],"related":[{"id":"z-ai/glm-5.3","name":"GLM-5.3","benchmarkUrl":"/benchmarks/models/glm-5.3"}],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"},{"label":"GLM-5.3-Flash-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF"}]},"deploymentClass":{"value":"multi-GPU workstation/server","status":"inferred","sources":[{"label":"GLM-5.3-Flash","url":"https://huggingface.co/zai-org/GLM-5.3-Flash"},{"label":"GLM-5.3-Flash-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF"}]},"weightFootprints":[{"artifact":"Unsloth UD-Q4_K_XL GGUF","precisionOrQuantization":"~4-bit","sizeGB":200,"status":"verified","sources":[{"label":"GLM-5.3-Flash-GGUF","url":"https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF"}]}],"componentStorage":{}},"special":"Formerly Ox Alpha, this MIT-licensed multimodal MoE combines a million-token context with low hosted token rates.","bestUseCase":"Cost-conscious coding and multimodal agents through an API, or self-hosting where several hundred gigabytes of model memory are available.","pricingSummary":{"value":"Vercel AI Gateway / DigitalOcean route; input 0.15 USD/1M; cached read 0.03/1M; output 0.5/1M; A specific gateway provider route; Vercel lists other providers with different prices and limits.","status":"verified","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"speedSummary":{"value":"Vercel AI Gateway / DigitalOcean: 119 output tokens/s, observed 2026-10-06.","status":"reviewed","sources":[{"label":"GLM 5.3 Flash API, Pricing & Playground","url":"https://vercel.com/ai-gateway/models/glm-5.3-flash"}]},"relatedArtifacts":[],"normalTokensPerSecond":88.98,"normalTokensPerSecondMin":18.65,"normalTokensPerSecondMax":127.37},"owl-alpha":{"id":"owl-alpha","name":"LongCat-2.0","facts":{"family":{"value":"LongCat","status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},"developer":{"value":"Meituan","status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},"origin":{"value":"China","status":"verified","sources":[{"label":"Investor Relations","url":"https://www.meituan.com/en-US/investor-relations"}]},"license":{"value":"MIT","status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}],"url":"https://huggingface.co/meituan-longcat/LongCat-2.0/blob/main/LICENSE"},"architecture":{"value":"Mixture-of-Experts with LongCat Sparse Attention","status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},"contextWindowTokens":{"value":1000000,"status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},"inputModalities":{"value":["text"],"status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},"outputModalities":{"value":["text"],"status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},"weightsAvailability":{"value":"official-open","status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},"activeParameters":{"value":48000000000,"status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]}},"access":[{"kind":"router","label":"OpenRouter model page","url":"https://openrouter.ai/meituan/longcat-2.0","providerModelId":"meituan/longcat-2.0"},{"kind":"api","label":"OpenRouter Chat Completions","url":"https://openrouter.ai/api/v1/chat/completions","providerModelId":"meituan/longcat-2.0"},{"kind":"download","label":"Official Hugging Face weights","url":"https://huggingface.co/meituan-longcat/LongCat-2.0","providerModelId":"meituan-longcat/LongCat-2.0","verifiedAt":"2026-10-06","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"}]},{"kind":"api","label":"LongCat API","url":"https://longcat.chat/platform/docs/Pricing/LongCat-2.0.html","providerModelId":"LongCat-2.0","verifiedAt":"2026-10-06","sources":[{"label":"LongCat-2.0 pricing","url":"https://longcat.chat/platform/docs/Pricing/LongCat-2.0.html"}]}],"pricing":[{"provider":"OpenRouter","providerModelId":"meituan/longcat-2.0","currency":"USD","unit":"per-1M-tokens","input":0.3,"output":1.2,"cachedRead":0.006,"fetchedAt":"2026-10-23T20:44:15.599Z","observedAt":"2026-10-23T20:44:15.599Z","status":"verified","sources":[{"label":"OpenRouter model pricing feed","url":"https://openrouter.ai/api/v1/models"}]},{"provider":"LongCat API","providerModelId":"LongCat-2.0","observedAt":"2026-10-06","currency":"USD","unit":"per-1M-tokens","input":0.3,"output":1.2,"cachedRead":0.006,"availability":"active","status":"verified","sources":[{"label":"LongCat-2.0 pricing","url":"https://longcat.chat/platform/docs/Pricing/LongCat-2.0.html"}]}],"speed":[],"related":[],"deployment":{"selfHostable":{"value":true,"status":"verified","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"},{"label":"LongCat 2.0 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/longcat-2-0"}]},"deploymentClass":{"value":"rack-scale","status":"inferred","sources":[{"label":"LongCat-2.0","url":"https://huggingface.co/meituan-longcat/LongCat-2.0"},{"label":"LongCat 2.0 - Intelligence, Performance & Price Analysis","url":"https://artificialanalysis.ai/models/longcat-2-0"}]},"weightFootprints":[],"componentStorage":{}},"special":"Revealed from Owl Alpha, LongCat-2.0 provides MIT-licensed MoE weights, sparse attention and a million-token context.","bestUseCase":"Long-context coding and agent workflows through hosted access, or deployment on large GPU clusters.","pricingSummary":{"value":"LongCat API; input 0.3 USD/1M; cached read 0.006/1M; output 1.2/1M; Limited-time discounted USD price; list price is $0.75 input / $0.015 cached input / $2.95 output. Expiry date not stated on page.","status":"verified","sources":[{"label":"LongCat-2.0 pricing","url":"https://longcat.chat/platform/docs/Pricing/LongCat-2.0.html"}]},"relatedArtifacts":[],"normalTokensPerSecond":86.39,"normalTokensPerSecondMin":84.11,"normalTokensPerSecondMax":88.44},"space-bunny-alpha":{"id":"space-bunny-alpha","name":"Space Bunny Alpha","normalTokensPerSecond":299.71,"normalTokensPerSecondMin":277.34,"normalTokensPerSecondMax":322.08},"fledge-alpha":{"id":"fledge-alpha","name":"Fledge Alpha","normalTokensPerSecond":62.71,"normalTokensPerSecondMin":60.52,"normalTokensPerSecondMax":70.8},"glyph-cluster":{"id":"glyph-cluster","name":"Glyph Cluster","normalTokensPerSecond":68.68,"normalTokensPerSecondMin":67.08,"normalTokensPerSecondMax":70.26},"missingno":{"id":"missingno","name":"MISSINGNO.","normalTokensPerSecond":845.75,"normalTokensPerSecondMin":823.99,"normalTokensPerSecondMax":858.88},"iceman-contributor":{"id":"iceman-contributor","name":"Iceman Contributor","normalTokensPerSecond":46.29,"normalTokensPerSecondMin":18.07,"normalTokensPerSecondMax":61.69}}}