{
  "slug": "olmo-2-7b",
  "order": 3,
  "name": "OLMo 2 (7B)",
  "baseTitle": "OLMo 2 7B",
  "company": "Ai2",
  "country": "United States",
  "flag": "🇺🇸",
  "date": "2024-11-25",
  "decoderType": "Dense",
  "attention": "MHA with QK-Norm",
  "layerMix": "32 MHA",
  "scale": "7B parameters",
  "totalParams": 7000000000,
  "activeParams": 7000000000,
  "context": "4,096",
  "contextTokens": 4096,
  "license": "Apache License 2.0",
  "kvCacheText": "512 KiB",
  "kvCacheKiB": 512,
  "kvCacheLevel": "Very high",
  "summary": "Transparent dense model that keeps classic MHA and pushes normalization changes for training stability.",
  "keyDetail": "Uses inside-residual post-norm instead of the usual pre-norm layout.",
  "image": {
    "full": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/olmo-2-7b.webp",
    "thumb": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/thumbnails/olmo-2-7b.webp",
    "alt": "Figure 10: OLMo 2 7B",
    "width": 925,
    "height": 960
  },
  "articleUrl": "https://magazine.sebastianraschka.com/p/the-big-llm-architecture-comparison#%C2%A723-olmo-2-summary",
  "links": [
    {
      "label": "View in article",
      "href": "https://magazine.sebastianraschka.com/p/the-big-llm-architecture-comparison#%C2%A723-olmo-2-summary",
      "title": null
    },
    {
      "label": "config.json",
      "href": "https://huggingface.co/allenai/OLMo-2-1124-7B-Instruct/blob/main/config.json",
      "title": "allenai/OLMo-2-1124-7B-Instruct"
    },
    {
      "label": "Tech report",
      "href": "https://arxiv.org/pdf/2501.00656",
      "title": null
    }
  ],
  "concepts": [
    {
      "slug": "qk-norm",
      "label": "QK-Norm",
      "name": "QK-Norm",
      "url": "https://llm-atlas-7or.pages.dev/concepts/qk-norm/"
    },
    {
      "slug": "mha",
      "label": "MHA",
      "name": "Multi-Head Attention",
      "url": "https://llm-atlas-7or.pages.dev/concepts/mha/"
    }
  ],
  "aai": {
    "v4": {
      "label": "AA Intelligence Index v4.3.2",
      "total": 5.6,
      "profile": {},
      "url": "https://artificialanalysis.ai/models/olmo-2-7b"
    }
  },
  "sourceUrl": "https://sebastianraschka.com/llm-architecture-gallery/#card-olmo-2-7b",
  "family": "dense",
  "url": "https://llm-atlas-7or.pages.dev/models/olmo-2-7b/",
  "markdownUrl": "https://llm-atlas-7or.pages.dev/models/olmo-2-7b.md"
}