{"id":"model.mistral.small","name":"Mistral Small","family":"Mistral Small","type":"frontier","modalities":["text","image"],"capabilities":["agent","reasoning","tool_calls","vision"],"openness":{"weights":"open_weights","code":"not_disclosed","data":"not_disclosed","license":null,"deployment":"API_and_self_deployment","api":null},"keySpecs":{"contextWindowTokens":256000,"extendedContextWindowTokens":null,"totalParameters":119,"activeParameters":null,"architectureType":null,"baseModel":null,"inputPriceUsdPerMtok":null,"outputPriceUsdPerMtok":null,"concurrencyLimitRequests":null},"organization":{"id":"org.mistral","name":"Mistral AI","websiteUrl":"https://mistral.ai"},"brand":{"id":"brand.mistral","name":"Mistral","websiteUrl":"https://mistral.ai"},"product":{"id":"product.mistral.api","name":"Mistral API","websiteUrl":"https://docs.mistral.ai/"},"origin":"international","versions":[{"id":"version.mistral.small-3-2","name":"Mistral Small 3.2","externalModelId":"mistral-small-2506","kind":"date","releaseDate":"2025-06-20","datePrecision":"day","publicAvailableAt":null,"lifecycleStatus":"deprecated","knowledgeCutoff":null,"claims":[],"benchmarks":[]},{"id":"version.mistral.small-4","name":"Mistral Small 4","externalModelId":"mistral-small-2603","kind":"date","releaseDate":"2026-03-16","datePrecision":"day","publicAvailableAt":null,"lifecycleStatus":"active","knowledgeCutoff":null,"claims":[{"id":"claim.version.mistral.small-4.capability","fieldKey":"capability_tags","value":{"value":["agent","vision","reasoning","tool_calls"]},"rawValue":null,"ingestionOrigin":"curated_official","unit":null,"condition":"Set on 2026-10-20 from the model card: agentic capabilities with native function calling and JSON output, image input support, and a togglable reasoning mode.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}},{"id":"claim.version.mistral.small-4.hf.context","fieldKey":"context_window_tokens","value":{"unit":"tokens","value":256000},"rawValue":null,"ingestionOrigin":"curated_official","unit":"tokens","condition":"Re-verified on 2026-10-20 from the model card: 256k context length.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}},{"id":"claim.version.mistral.small-4.deployment","fieldKey":"deployment_options","value":{"value":"self_hosted 119B MoE weights; vLLM tag; NVFP4 quantised checkpoint; 4x NVIDIA HGX H100 minimum"},"rawValue":null,"ingestionOrigin":"curated_official","unit":null,"condition":"Re-verified on 2026-10-20; the card is tagged vLLM and provides NVFP4 and eagle-head checkpoints. The minimum 4x H100 footprint was recorded earlier from the release announcement and is not restated on the model card.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}},{"id":"claim.version.mistral.small-4.lifecycle-status","fieldKey":"lifecycle_status","value":{"value":"active"},"rawValue":null,"ingestionOrigin":"curated_official","unit":null,"condition":"Set on 2026-10-20 because the mistralai Hugging Face repository for Small 4 remains public, not gated and not disabled; Mistral does not publish a formal lifecycle table for these weights, so this reflects current availability rather than a vendor-stated status.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}},{"id":"claim.version.mistral.small-4.license","fieldKey":"model_license","value":{"value":"Apache 2.0"},"rawValue":null,"ingestionOrigin":"curated_official","unit":null,"condition":"Re-verified on 2026-10-20; the card states Apache 2.0 for commercial and non-commercial use and front matter declares apache-2.0.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}},{"id":"claim.version.mistral.small-4.hf.release-date","fieldKey":"release_date","value":{"value":"2026-03","precision":"month"},"rawValue":null,"ingestionOrigin":"curated_official","unit":null,"condition":"Set on 2026-10-20 from the official repository name suffix 2603; the model card does not state a day, so only month precision is asserted here.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}},{"id":"claim.version.mistral.small-4.hf.total-parameters","fieldKey":"total_parameters","value":{"unit":"B","value":119},"rawValue":null,"ingestionOrigin":"curated_official","unit":"B","condition":"Re-verified on 2026-10-20 from the model card: 119B parameters with 6.5B activated per token.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}},{"id":"claim.version.mistral.small-4.weights","fieldKey":"weights_availability","value":{"value":"open_weights"},"rawValue":null,"ingestionOrigin":"curated_official","unit":null,"condition":"Re-verified on 2026-10-20 against the official Mistral AI Hugging Face repository, which publishes the Small 4 weights, NVFP4 quantisation and an eagle head and is not gated.","status":"approved","confidence":"high","reviewedBy":"system:curated-publisher","reviewedAt":"2026-10-20 11:11:33.691284+08","effectiveFrom":null,"effectiveTo":null,"source":{"id":"source.mistral.hf.small-4","title":"Mistral Small 4 119B A6B (official model card)","url":"https://huggingface.co/mistralai/Mistral-Small-4-119B-2603","accessedAt":"2026-10-20 11:11:14.035948+08","robotsStatus":"allowed","collectionEligibility":"eligible","collectionCheckedAt":"2026-10-20 11:11:14.035948+08"},"evidence":{"id":"evidence.mistral.small-4.hf.2026-10-20","locator":"Hugging Face model card mistralai/Mistral-Small-4-119B-2603 (README.md), Key Features section and front-matter license","excerpt":"Mistral Small 4 is a hybrid model that unifies Instruct, Reasoning and Devstral capabilities into one multimodal model. Key features: MoE with 128 experts and 4 active; 119B parameters with 6.5B activated per token; 256k context length; multimodal input accepting text and images with text output; instruct and reasoning functionality with function calls and configurable reasoning effort; vision, multilingual and agentic capabilities with native function calling and JSON output; Apache 2.0 license for commercial and non-commercial use. The card is tagged vLLM and links NVFP4 and eagle speculative-decoding checkpoints. Front matter: license: apache-2.0. The Hugging Face model API reports the repository as public and not gated."}}],"benchmarks":[]}],"dataUpdatedAt":"2026-10-20 11:11:33.691284+08","latestVersion":{"id":"version.mistral.small-4","name":"Mistral Small 4","externalModelId":"mistral-small-2603","status":"active"},"dataQuality":{"approved":8,"pending":0,"stale":17,"conflicted":1},"availabilityRegions":[],"availabilityChannels":[],"availabilityScopes":[],"decisionReadiness":{"overall":"A","evaluatedAt":"2026-11-01T12:03:35.113Z","tracks":[{"key":"api","applicable":false,"grade":"incomplete","verifiedAt":null,"satisfiedFields":[],"missingFields":[],"missingEvidenceFields":[],"staleFields":[],"conflictedFields":[],"ineligibleSourceFields":[],"reasons":[],"fields":[]},{"key":"self_hosted","applicable":true,"grade":"A","verifiedAt":"2026-10-20 11:11:33.691284+08","satisfiedFields":["release_date","lifecycle_status","capability_tags","deployment_options","context_window_tokens","weights_availability","model_license","total_parameters"],"missingFields":[],"missingEvidenceFields":[],"staleFields":[],"conflictedFields":[],"ineligibleSourceFields":[],"reasons":[],"fields":[{"fieldKey":"release_date","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2027-01-18T03:11:33.691Z"},{"fieldKey":"lifecycle_status","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2026-11-19T03:11:33.691Z"},{"fieldKey":"capability_tags","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2027-01-18T03:11:33.691Z"},{"fieldKey":"deployment_options","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2027-01-18T03:11:33.691Z"},{"fieldKey":"context_window_tokens","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2027-01-18T03:11:33.691Z"},{"fieldKey":"weights_availability","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2026-11-19T03:11:33.691Z"},{"fieldKey":"model_license","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2027-01-18T03:11:33.691Z"},{"fieldKey":"total_parameters","status":"satisfied","verifiedAt":"2026-10-20 11:11:33.691284+08","expiresAt":"2027-01-18T03:11:33.691Z"}]}]},"timeline":[{"id":"timeline.version.version.mistral.small-3-2","kind":"version","subjectId":"version.mistral.small-3-2","date":"2025-06-20","datePrecision":"day","title":"Mistral Small 3.2","summary":"模型版本 · deprecated","eventType":"version","lifecycleStatus":"deprecated","source":null},{"id":"timeline.version.version.mistral.small-4","kind":"version","subjectId":"version.mistral.small-4","date":"2026-03-16","datePrecision":"day","title":"Mistral Small 4","summary":"模型版本 · active","eventType":"version","lifecycleStatus":"active","source":null},{"id":"event.version.mistral.small-4.release-2026-03-16","kind":"event","subjectId":"version.mistral.small-4","date":"2026-03-16","datePrecision":"day","title":"Mistral Small 4 announcement","summary":"Mistral announced Small 4 as a unified model combining instruct, reasoning, multimodal and agentic coding capabilities.","eventType":"release","lifecycleStatus":null,"source":{"id":"source.mistral.news.small-4","title":"Introducing Mistral Small 4","url":"https://mistral.ai/news/mistral-small-4/"}}],"availability":[],"enterpriseCases":[],"useCases":[],"deploymentProfiles":[],"learningResources":[],"trainingProfiles":[],"publication":{"approvedClaimsOnly":true,"note":"页面只展示已审核 Claim；待审核数据不会被当作公开事实。"}}