{
  "slug": "olmo-3-32b",
  "order": 25,
  "name": "OLMo 3 (32B)",
  "baseTitle": "OLMo 3 32B",
  "company": "Ai2",
  "country": "United States",
  "flag": "🇺🇸",
  "date": "2025-11-20",
  "decoderType": "Dense",
  "attention": "GQA with QK-Norm and 3:1 sliding-window/global attention",
  "layerMix": "48 sliding-window + 16 global",
  "scale": "32B parameters",
  "totalParams": 32000000000,
  "activeParams": 32000000000,
  "context": "65,536",
  "contextTokens": 65536,
  "license": "Apache License 2.0",
  "kvCacheText": "256 KiB",
  "kvCacheKiB": 256,
  "kvCacheLevel": "High",
  "summary": "Scaled-up OLMo 3 keeps the same block design but moves to grouped-query attention.",
  "keyDetail": "Keeps post-norm while scaling width and applying YaRN only on global layers.",
  "image": {
    "full": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/olmo-3-32b.webp",
    "thumb": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/thumbnails/olmo-3-32b.webp",
    "alt": "Figure 45: OLMo 3 32B",
    "width": 924,
    "height": 960
  },
  "articleUrl": "https://magazine.sebastianraschka.com/p/the-big-llm-architecture-comparison#%C2%A715-olmo-3-thinking",
  "links": [
    {
      "label": "View in article",
      "href": "https://magazine.sebastianraschka.com/p/the-big-llm-architecture-comparison#%C2%A715-olmo-3-thinking",
      "title": null
    },
    {
      "label": "From scratch",
      "href": "https://github.com/rasbt/LLMs-from-scratch/tree/main/ch05/13_olmo3",
      "title": null
    },
    {
      "label": "config.json",
      "href": "https://huggingface.co/allenai/Olmo-3-32B-Think/blob/main/config.json",
      "title": "allenai/Olmo-3-32B-Think"
    },
    {
      "label": "Tech report",
      "href": "https://arxiv.org/pdf/2512.13961",
      "title": null
    }
  ],
  "concepts": [
    {
      "slug": "qk-norm",
      "label": "QK-Norm",
      "name": "QK-Norm",
      "url": "https://llm-atlas-7or.pages.dev/concepts/qk-norm/"
    },
    {
      "slug": "gqa",
      "label": "GQA",
      "name": "Grouped-Query Attention",
      "url": "https://llm-atlas-7or.pages.dev/concepts/gqa/"
    },
    {
      "slug": "swa",
      "label": "SWA",
      "name": "Sliding Window Attention",
      "url": "https://llm-atlas-7or.pages.dev/concepts/swa/"
    }
  ],
  "aai": {
    "v4": {
      "label": "AA Intelligence Index v4.3.2",
      "total": 6.5,
      "profile": {
        "Scientific": 3.2
      },
      "url": "https://artificialanalysis.ai/models/olmo-3-32b-think"
    }
  },
  "sourceUrl": "https://sebastianraschka.com/llm-architecture-gallery/#card-olmo-3-32b",
  "family": "dense",
  "url": "https://llm-atlas-7or.pages.dev/models/olmo-3-32b/",
  "markdownUrl": "https://llm-atlas-7or.pages.dev/models/olmo-3-32b.md"
}