{
  "slug": "mistral-small-4",
  "order": 55,
  "name": "Mistral Small 4 (119B)",
  "baseTitle": "Mistral Small 4",
  "company": "Mistral AI",
  "country": "France",
  "flag": "🇫🇷",
  "date": "2026-03-16",
  "decoderType": "Sparse MoE",
  "attention": "MLA",
  "layerMix": "36 MLA",
  "scale": "119B total, 6.63B active (5.6% active)",
  "totalParams": 119000000000,
  "activeParams": 6630000000,
  "context": "256,000",
  "contextTokens": 256000,
  "license": "Apache License 2.0",
  "kvCacheText": "22.5 KiB",
  "kvCacheKiB": 22.5,
  "kvCacheLevel": "Very low",
  "summary": "Multimodal Mistral Small refresh that jumps from the older dense 24B stack to an MLA-based sparse MoE design.",
  "keyDetail": "Uses 128 experts with 4 routed plus 1 shared expert active per token while unifying instruct, reasoning, and vision.",
  "image": {
    "full": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/mistral-small-4.webp",
    "thumb": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/thumbnails/mistral-small-4.webp",
    "alt": "Figure 1: Mistral Small 4",
    "width": 960,
    "height": 938
  },
  "articleUrl": null,
  "links": [
    {
      "label": "config.json",
      "href": "https://huggingface.co/mistralai/Mistral-Small-4-119B-2603/blob/main/config.json",
      "title": "mistralai/Mistral-Small-4-119B-2603"
    },
    {
      "label": "Tech report",
      "href": "https://mistral.ai/news/mistral-small-4",
      "title": null
    }
  ],
  "concepts": [
    {
      "slug": "mla",
      "label": "MLA",
      "name": "Multi-Head Latent Attention",
      "url": "https://llm-atlas-7or.pages.dev/concepts/mla/"
    },
    {
      "slug": "moe",
      "label": "MoE",
      "name": "Mixture of Experts",
      "url": "https://llm-atlas-7or.pages.dev/concepts/moe/"
    }
  ],
  "aai": {
    "v5": {
      "label": "AA Intelligence Index v5.0",
      "total": 4.3,
      "profile": {
        "Knowledge Work": 0.2,
        "Coding": 0,
        "Science": 0,
        "General": 17
      },
      "url": "https://artificialanalysis.ai/models/mistral-small-4"
    },
    "v4": {
      "label": "AA Intelligence Index v4.3.2",
      "total": 11.3,
      "profile": {
        "General": 21.1,
        "Scientific": 5.1,
        "Coding": 19.4,
        "Agents": 0.2
      },
      "url": "https://artificialanalysis.ai/models/mistral-small-4"
    }
  },
  "sourceUrl": "https://sebastianraschka.com/llm-architecture-gallery/#card-mistral-small-4",
  "family": "moe",
  "url": "https://llm-atlas-7or.pages.dev/models/mistral-small-4/",
  "markdownUrl": "https://llm-atlas-7or.pages.dev/models/mistral-small-4.md"
}