enosys/gigachat3.1-10b-a1.8b-750-v1-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "mit",
    "base_model": [
      "ai-sage/GigaChat3.1-10B-A1.8B-bf16"
    ],
    "datasets": [
      "eaddario/imatrix-calibration"
    ],
    "language": [
      "ru",
      "en"
    ],
    "pipeline_tag": "text-generation",
    "tags": [
      "moe"
    ],
    "frontmatter": {
      "license": "mit",
      "base_model": [
        "ai-sage/GigaChat3.1-10B-A1.8B-bf16"
      ],
      "datasets": [
        "eaddario/imatrix-calibration"
      ],
      "language": [
        "ru",
        "en"
      ],
      "pipeline_tag": "text-generation",
      "tags": [
        "moe"
      ]
    },
    "hero_image_url": "",
    "summary": "Many thanks to Ed Addario for an impressive job.",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: mit\nbase_model:\n- ai-sage/GigaChat3.1-10B-A1.8B-bf16\ndatasets:\n- eaddario/imatrix-calibration\nlanguage:\n- ru\n- en\npipeline_tag: text-generation\ntags:\n- moe\n---\n\n# Experimental global target bits‑per‑weight quantization of ai-sage/GigaChat3.1-10B-A1.8B-bf16\n- Using **non-standard** (forked) [LLaMA C++](https://github.com/EAddario/llama.cpp/tree/quantize) branch for quantization.\n- Using a [CLI tool](https://github.com/cmhamiche/kld-sweep-dataset) to build KLD evaluation and imatrix calibration datasets for GGUF models, sourced from [eaddario/imatrix-calibration](https://huggingface.co/datasets/eaddario/imatrix-calibration).\n- Using dataset sources: tools, math, code, text_en, text_ru.\n- Using dataset chunks: 750.\n- Tensors quantinization F16 instead of BF16, Nvidia Pascal architecture friendly like P100.\n- Small set of patches added.\n\nMany thanks to [Ed Addario](https://huggingface.co/eaddario) for an impressive job.\n\n## Quantization comparison\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "moe",
    "text-generation",
    "ru",
    "en",
    "dataset:eaddario/imatrix-calibration",
    "base_model:ai-sage/GigaChat3.1-10B-A1.8B-bf16",
    "base_model:quantized:ai-sage/GigaChat3.1-10B-A1.8B-bf16",
    "license:mit",
    "endpoints_compatible",
    "region:us",
    "imatrix",
    "conversational"
  ],
  "likes": 1,
  "downloads": 271,
  "gated": false,
  "private": false,
  "last_modified": "2026-04-17T12:59:54.000Z",
  "created_at": "2026-04-13T08:07:13.000Z",
  "pipeline_tag": "text-generation",
  "library_name": ""
}

{
  "_id": "69dca431802d89435c5d34d1",
  "id": "ENOSYS/GigaChat3.1-10B-A1.8B-750-v1-GGUF",
  "modelId": "ENOSYS/GigaChat3.1-10B-A1.8B-750-v1-GGUF",
  "sha": "f1b4cb4ab4283a3247eac8ccadc0f1fb7eae171a",
  "createdAt": "2026-04-13T08:07:13.000Z",
  "lastModified": "2026-04-17T12:59:54.000Z",
  "author": "ENOSYS",
  "downloads": 271,
  "likes": 1,
  "gated": false,
  "private": false,
  "pipeline_tag": "text-generation",
  "library_name": "",
  "siblings_count": 142
}