{
  "data": {
    "tool": {
      "author": "google",
      "category": "inference",
      "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/gemma.json",
      "kind": "model",
      "listed": "indexed",
      "maker": "Google",
      "markdownUrl": "https://www.anchorterminal.com/tools/gemma.md",
      "modelCount": 6,
      "models": [
        {
          "contextTokens": 262144,
          "freeVariant": true,
          "id": "google/gemma-4-26b-a4b-it",
          "input": [
            "image",
            "text",
            "video"
          ],
          "inputUsdPerM": 0.0675,
          "maxOutput": 235929,
          "name": "Gemma 4 26B A4B",
          "outputUsdPerM": 0.225,
          "promptCaching": true,
          "reasoning": true,
          "released": "2026-04-03",
          "structuredOutputs": true,
          "tools": true
        },
        {
          "contextTokens": 262144,
          "freeVariant": true,
          "id": "google/gemma-4-31b-it",
          "input": [
            "image",
            "text",
            "video"
          ],
          "inputUsdPerM": 0.09,
          "maxOutput": 16384,
          "name": "Gemma 4 31B",
          "outputUsdPerM": 0.34,
          "promptCaching": true,
          "reasoning": true,
          "released": "2026-04-02",
          "structuredOutputs": true,
          "tools": true
        },
        {
          "contextTokens": 131072,
          "id": "google/gemma-3-4b-it",
          "input": [
            "image",
            "text"
          ],
          "inputUsdPerM": 0.05,
          "maxOutput": 16384,
          "name": "Gemma 3 4B",
          "outputUsdPerM": 0.1,
          "promptCaching": false,
          "reasoning": false,
          "released": "2025-03-13",
          "structuredOutputs": true,
          "tools": false
        },
        {
          "contextTokens": 131072,
          "id": "google/gemma-3-12b-it",
          "input": [
            "image",
            "text"
          ],
          "inputUsdPerM": 0.05,
          "maxOutput": 16384,
          "name": "Gemma 3 12B",
          "outputUsdPerM": 0.15,
          "promptCaching": false,
          "reasoning": false,
          "released": "2025-03-13",
          "structuredOutputs": true,
          "tools": true
        },
        {
          "contextTokens": 131072,
          "id": "google/gemma-3-27b-it",
          "input": [
            "image",
            "text"
          ],
          "inputUsdPerM": 0.08,
          "maxOutput": 117964,
          "name": "Gemma 3 27B",
          "outputUsdPerM": 0.45,
          "promptCaching": true,
          "reasoning": false,
          "released": "2025-03-12",
          "structuredOutputs": true,
          "tools": true
        },
        {
          "contextTokens": 8192,
          "id": "google/gemma-2-27b-it",
          "input": [
            "text"
          ],
          "inputUsdPerM": 0.65,
          "maxOutput": 2048,
          "name": "Gemma 2 27B",
          "outputUsdPerM": 0.65,
          "promptCaching": false,
          "reasoning": false,
          "released": "2024-07-13",
          "structuredOutputs": true,
          "tools": false
        }
      ],
      "name": "Gemma",
      "note": "Indexed from OpenRouter: facts and our own checks, not reviewed, so no score, grade or rank.",
      "pageJsonUrl": "https://www.anchorterminal.com/tools/gemma.json",
      "reviewed": false,
      "slug": "gemma",
      "source": "OpenRouter",
      "sourceUrl": "https://openrouter.ai/google/gemma-4-26b-a4b-it",
      "summary": "Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...",
      "updatedAt": "2026-04-03T14:53:09Z",
      "url": "https://www.anchorterminal.com/tools/gemma",
      "vendor": "Google",
      "vendorUrl": "",
      "why": [
        "openrouter"
      ]
    }
  },
  "kind": "anchor.page",
  "links": {
    "api": "https://www.anchorterminal.com/api/v1/index.json",
    "html": "https://www.anchorterminal.com/tools/gemma",
    "json": "https://www.anchorterminal.com/tools/gemma.json",
    "llms": "https://www.anchorterminal.com/llms.txt",
    "markdown": "https://www.anchorterminal.com/tools/gemma.md",
    "slim": "https://www.anchorterminal.com/tools/gemma.min.md"
  },
  "markdown": "# Gemma\n\n\u003e Indexed, not reviewed: facts from OpenRouter and our own checks. No score, grade or rank, and not in the rankings until the panel reviews it. How the index works: https://www.anchorterminal.com/indexed/\n\n- Kind: model family, by Google\n- Category: Model APIs \u0026 inference (https://www.anchorterminal.com/categories/inference.md)\n- Listed because: OpenRouter lists it, with prices from the providers that serve it.\n- What OpenRouter says: Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...\n\n## Models (6)\n\n| Model | OpenRouter id | Released | Context | Input $/1M | Output $/1M | Supports |\n| --- | --- | --- | --- | --- | --- | --- |\n| Gemma 4 26B A4B | `google/gemma-4-26b-a4b-it` | 2026-04 | 262k | $0.0675 | $0.225 | tools, structured output, reasoning, caching, image, video, free variant |\n| Gemma 4 31B | `google/gemma-4-31b-it` | 2026-04 | 262k | $0.09 | $0.34 | tools, structured output, reasoning, caching, image, video, free variant |\n| Gemma 3 4B | `google/gemma-3-4b-it` | 2025-03 | 131k | $0.05 | $0.10 | structured output, image |\n| Gemma 3 12B | `google/gemma-3-12b-it` | 2025-03 | 131k | $0.05 | $0.15 | tools, structured output, image |\n| Gemma 3 27B | `google/gemma-3-27b-it` | 2025-03 | 131k | $0.08 | $0.45 | tools, structured output, caching, image |\n| Gemma 2 27B | `google/gemma-2-27b-it` | 2024-07 | 8k | $0.65 | $0.65 | structured output |\n\nPrices and limits as OpenRouter lists them; a provider's own price can differ.\n\n- JSON: https://www.anchorterminal.com/api/v1/tools/gemma.json\n- Being indexed says nothing about quality, and nobody can pay for it. Ask for a review: https://www.anchorterminal.com/builders/#claiming\n",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-05",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.3",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  },
  "page": {
    "breadcrumbs": [
      {
        "name": "Home",
        "url": "https://www.anchorterminal.com/"
      },
      {
        "name": "Terminal",
        "url": "https://www.anchorterminal.com/tools/"
      },
      {
        "name": "Indexed",
        "url": "https://www.anchorterminal.com/indexed/"
      },
      {
        "name": "Gemma",
        "url": ""
      }
    ],
    "description": "Gemma, a family of models by Google, listed from OpenRouter. Indexed, not reviewed: facts and our own checks, no score or ranking. Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate…",
    "facts": [
      "not reviewed",
      "not ranked",
      "facts only"
    ],
    "h1": "Gemma",
    "image": "https://www.anchorterminal.com/assets/og/indexed.png",
    "path": "/tools/gemma",
    "published": "",
    "section": "indexed",
    "title": "Gemma: family of models, indexed from OpenRouter | Anchor Terminal",
    "toc": null,
    "updated": "2026-10-05",
    "url": "https://www.anchorterminal.com/tools/gemma"
  },
  "tokens": {
    "markdown": 650,
    "slim": 580
  },
  "version": 1
}