{
  "slug": "deepseek-v4-1-flash",
  "order": 102,
  "name": "DeepSeek V4.1-Flash (552B)",
  "baseTitle": "DeepSeek V4.1-Flash",
  "company": "DeepSeek",
  "country": "China",
  "flag": "🇨🇳",
  "date": "2026-09-10",
  "decoderType": "Sparse MoE (causal encoder-decoder)",
  "attention": "MLA-style CSA2 + SWA with cross-layer KV and sparse-index reuse",
  "layerMix": "20 encoder + 20 decoder layers; 2 SWA-only + 38 CSA2/SWA",
  "scale": "552B backbone + 196B Engram; 8B active prefill, 16B active decode (2.9% of backbone)",
  "totalParams": 552000000000,
  "activeParams": 8000000000,
  "context": "1,048,576",
  "contextTokens": 1048576,
  "license": "MIT License",
  "kvCacheText": "3.125 KiB",
  "kvCacheKiB": 3.125,
  "kvCacheLevel": "Very low",
  "summary": "Causal encoder-decoder MoE with CSA2 cache sharing, Engram memory, and a million-token context. The figure shows the text backbone.",
  "keyDetail": "Decoder global KV comes from final encoder states. Single-Pass mHC mixes four residual streams around each attention and MoE sublayer.",
  "image": {
    "full": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/deepseek-v4-1-flash.webp",
    "thumb": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/thumbnails/deepseek-v4-1-flash.webp",
    "alt": "Figure 1: DeepSeek V4.1-Flash",
    "width": 960,
    "height": 683
  },
  "articleUrl": null,
  "links": [
    {
      "label": "config.json",
      "href": "https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/main/config.json",
      "title": "deepseek-ai/DeepSeek-V4.1-Flash"
    },
    {
      "label": "License",
      "href": "https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/main/LICENSE",
      "title": null
    },
    {
      "label": "Tech report",
      "href": "https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/main/DeepSeek_V41_Tech_Report.pdf",
      "title": null
    }
  ],
  "concepts": [
    {
      "slug": "mhc",
      "label": "mHC",
      "name": "Manifold-constrained Hyper-Connections",
      "url": "https://llm-atlas-7or.pages.dev/concepts/mhc/"
    },
    {
      "slug": "kv-sharing",
      "label": "KV Sharing",
      "name": "Cross-Layer KV Sharing",
      "url": "https://llm-atlas-7or.pages.dev/concepts/kv-sharing/"
    },
    {
      "slug": "swa",
      "label": "SWA",
      "name": "Sliding Window Attention",
      "url": "https://llm-atlas-7or.pages.dev/concepts/swa/"
    },
    {
      "slug": "moe",
      "label": "MoE",
      "name": "Mixture of Experts",
      "url": "https://llm-atlas-7or.pages.dev/concepts/moe/"
    }
  ],
  "aai": {
    "v4": {
      "label": "AA Intelligence Index v4.3.2",
      "total": 39.5,
      "profile": {
        "General": 34.3,
        "Scientific": 26.8,
        "Coding": 39.3,
        "Agents": 52.8
      },
      "url": "https://artificialanalysis.ai/models/deepseek-v4-1-flash"
    }
  },
  "sourceUrl": "https://sebastianraschka.com/llm-architecture-gallery/#card-deepseek-v4-1-flash",
  "family": "moe",
  "url": "https://llm-atlas-7or.pages.dev/models/deepseek-v4-1-flash/",
  "markdownUrl": "https://llm-atlas-7or.pages.dev/models/deepseek-v4-1-flash.md"
}