{
  "slug": "sarvam-105b",
  "order": 48,
  "name": "Sarvam (105B)",
  "baseTitle": "Sarvam 105B",
  "company": "Sarvam AI",
  "country": "India",
  "flag": "🇮🇳",
  "date": "2026-03-03",
  "decoderType": "Sparse MoE",
  "attention": "MLA with KV LayerNorm and NoPE + RoPE",
  "layerMix": "32 MLA",
  "scale": "105B total, 10.3B active (9.8% active)",
  "totalParams": 105000000000,
  "activeParams": 10300000000,
  "context": "131,072",
  "contextTokens": 131072,
  "license": "Apache License 2.0",
  "kvCacheText": "36 KiB",
  "kvCacheKiB": 36,
  "kvCacheLevel": "Low",
  "summary": "Larger Sarvam variant keeps the sparse MoE layout but switches from GQA to MLA.",
  "keyDetail": "Large vocabulary and strong Indic language support carried into the larger MLA-based sparse MoE variant.",
  "image": {
    "full": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/sarvam-105b.webp",
    "thumb": "https://sebastianraschka.com/llm-architecture-gallery/images/architectures/thumbnails/sarvam-105b.webp",
    "alt": "Figure 38: Sarvam 105B",
    "width": 960,
    "height": 909
  },
  "articleUrl": "https://magazine.sebastianraschka.com/p/a-dream-of-spring-for-open-weight#%C2%A7update-1-sarvam-30b-and-105b-mar-6-2026",
  "links": [
    {
      "label": "View in article",
      "href": "https://magazine.sebastianraschka.com/p/a-dream-of-spring-for-open-weight#%C2%A7update-1-sarvam-30b-and-105b-mar-6-2026",
      "title": null
    },
    {
      "label": "config.json",
      "href": "https://huggingface.co/sarvamai/sarvam-105b/blob/main/config.json",
      "title": "sarvamai/sarvam-105b"
    },
    {
      "label": "Tech report",
      "href": "https://www.sarvam.ai/blogs/sarvam-30b-105b",
      "title": null
    }
  ],
  "concepts": [
    {
      "slug": "nope",
      "label": "NoPE",
      "name": "No Positional Embeddings",
      "url": "https://llm-atlas-7or.pages.dev/concepts/nope/"
    },
    {
      "slug": "gqa",
      "label": "GQA",
      "name": "Grouped-Query Attention",
      "url": "https://llm-atlas-7or.pages.dev/concepts/gqa/"
    },
    {
      "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": {
    "v4": {
      "label": "AA Intelligence Index v4.3.2",
      "total": 8.8,
      "profile": {
        "Scientific": 5.5
      },
      "url": "https://artificialanalysis.ai/models/sarvam-105b"
    }
  },
  "sourceUrl": "https://sebastianraschka.com/llm-architecture-gallery/#card-sarvam-105b",
  "family": "moe",
  "url": "https://llm-atlas-7or.pages.dev/models/sarvam-105b/",
  "markdownUrl": "https://llm-atlas-7or.pages.dev/models/sarvam-105b.md"
}