mradermacher/lfm2.5-1.2b-saiga-it-v2-i1-gguf overview

{
  "metadata": {},
  "card_data": {
    "base_model": "NickupAI/LFM2.5-1.2B-Saiga-It-v2",
    "datasets": [
      "IlyaGusev/saiga_scored",
      "d0rj/alpaca-cleaned-ru",
      "IlyaGusev/ru_sharegpt_cleaned",
      "IlyaGusev/ru_turbo_saiga",
      "lksy/ru_instruct_gpt4",
      "wikimedia/wikipedia",
      "uonlp/CulturaX",
      "allenai/c4"
    ],
    "language": [
      "ru",
      "en"
    ],
    "library_name": "transformers",
    "license": "other",
    "mradermacher": {
      "readme_rev": 1
    },
    "quantized_by": "mradermacher",
    "tags": [
      "russian",
      "saiga",
      "continued-pretraining",
      "sft",
      "lfm",
      "liquid-ai"
    ],
    "frontmatter": {
      "base_model": "NickupAI/LFM2.5-1.2B-Saiga-It-v2",
      "datasets": [
        "IlyaGusev/saiga_scored",
        "d0rj/alpaca-cleaned-ru",
        "IlyaGusev/ru_sharegpt_cleaned",
        "IlyaGusev/ru_turbo_saiga",
        "lksy/ru_instruct_gpt4",
        "wikimedia/wikipedia",
        "uonlp/CulturaX",
        "allenai/c4"
      ],
      "language": [
        "ru",
        "en"
      ],
      "library_name": "transformers",
      "license": "other",
      "mradermacher": [],
      "quantized_by": "mradermacher",
      "tags": [
        "russian",
        "saiga",
        "continued-pretraining",
        "sft",
        "lfm",
        "liquid-ai"
      ]
    },
    "hero_image_url": "https://www.nethype.de/huggingface_embed/quantpplgraph.png",
    "summary": "## About         weighted/imatrix quants of https://huggingface.co/NickupAI/LFM2.5-1.2B-Saiga-It-v2  ***For a convenient overview and download list, visit our model page for this model.*** static quants are available at https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-GGUF",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nbase_model: NickupAI/LFM2.5-1.2B-Saiga-It-v2\ndatasets:\n- IlyaGusev/saiga_scored\n- d0rj/alpaca-cleaned-ru\n- IlyaGusev/ru_sharegpt_cleaned\n- IlyaGusev/ru_turbo_saiga\n- lksy/ru_instruct_gpt4\n- wikimedia/wikipedia\n- uonlp/CulturaX\n- allenai/c4\nlanguage:\n- ru\n- en\nlibrary_name: transformers\nlicense: other\nmradermacher:\n  readme_rev: 1\nquantized_by: mradermacher\ntags:\n- russian\n- saiga\n- continued-pretraining\n- sft\n- lfm\n- liquid-ai\n---\n## About\n\n<!-- ### quantize_version: 2 -->\n<!-- ### output_tensor_quantised: 1 -->\n<!-- ### convert_type: hf -->\n<!-- ### vocab_type:  -->\n<!-- ### tags: nicoboss -->\n<!-- ### quants:  Q2_K IQ3_M Q4_K_S IQ3_XXS Q3_K_M small-IQ4_NL Q4_K_M IQ2_M Q6_K IQ4_XS Q2_K_S IQ1_M Q3_K_S IQ2_XXS Q3_K_L IQ2_XS Q5_K_S IQ2_S IQ1_S Q5_K_M Q4_0 IQ3_XS Q4_1 IQ3_S -->\n<!-- ### quants_skip:  -->\n<!-- ### skip_mmproj:  -->\nweighted/imatrix quants of https://huggingface.co/NickupAI/LFM2.5-1.2B-Saiga-It-v2\n\n<!-- provided-files -->\n\n***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#LFM2.5-1.2B-Saiga-It-v2-i1-GGUF).***\n\nstatic quants are available at https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-GGUF\n## Usage\n\nIf you are unsure how to use GGUF files, refer to one of [TheBloke's\nREADMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for\nmore details, including on how to concatenate multi-part files.\n\n## Provided Quants\n\n(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)\n\n| Link | Type | Size/GB | Notes |\n|:-----|:-----|--------:|:------|\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own quants) |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ1_S.gguf) | i1-IQ1_S | 0.4 | for the desperate |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ1_M.gguf) | i1-IQ1_M | 0.4 | mostly desperate |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 0.5 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ2_XS.gguf) | i1-IQ2_XS | 0.5 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ2_S.gguf) | i1-IQ2_S | 0.5 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ2_M.gguf) | i1-IQ2_M | 0.5 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q2_K_S.gguf) | i1-Q2_K_S | 0.6 | very low quality |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q2_K.gguf) | i1-Q2_K | 0.6 | IQ3_XXS probably better |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 0.6 | lower quality |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ3_XS.gguf) | i1-IQ3_XS | 0.6 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ3_S.gguf) | i1-IQ3_S | 0.7 | beats Q3_K* |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q3_K_S.gguf) | i1-Q3_K_S | 0.7 | IQ3_XS probably better |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ3_M.gguf) | i1-IQ3_M | 0.7 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q3_K_M.gguf) | i1-Q3_K_M | 0.7 | IQ3_S probably better |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q3_K_L.gguf) | i1-Q3_K_L | 0.7 | IQ3_M probably better |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ4_XS.gguf) | i1-IQ4_XS | 0.8 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-IQ4_NL.gguf) | i1-IQ4_NL | 0.8 | prefer IQ4_XS |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q4_0.gguf) | i1-Q4_0 | 0.8 | fast, low quality |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q4_K_S.gguf) | i1-Q4_K_S | 0.8 | optimal size/speed/quality |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q4_K_M.gguf) | i1-Q4_K_M | 0.8 | fast, recommended |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q4_1.gguf) | i1-Q4_1 | 0.9 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q5_K_S.gguf) | i1-Q5_K_S | 0.9 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q5_K_M.gguf) | i1-Q5_K_M | 0.9 |  |\n| [GGUF](https://huggingface.co/mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF/resolve/main/LFM2.5-1.2B-Saiga-It-v2.i1-Q6_K.gguf) | i1-Q6_K | 1.1 | practically like static Q6_K |\n\nHere is a handy graph by ikawrakow comparing some lower-quality quant\ntypes (lower is better):\n\n![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)\n\nAnd here are Artefact2's thoughts on the matter:\nhttps://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9\n\n## FAQ / Model Request\n\nSee https://huggingface.co/mradermacher/model_requests for some answers to\nquestions you might have and/or if you want some other model quantized.\n\n## Thanks\n\nI thank my company, [nethype GmbH](https://www.nethype.de/), for letting\nme use its servers and providing upgrades to my workstation to enable\nthis work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.\n\n<!-- end -->\n",
    "related_quantizations": []
  },
  "tags": [
    "transformers",
    "gguf",
    "russian",
    "saiga",
    "continued-pretraining",
    "sft",
    "lfm",
    "liquid-ai",
    "ru",
    "en",
    "dataset:IlyaGusev/saiga_scored",
    "dataset:d0rj/alpaca-cleaned-ru",
    "dataset:IlyaGusev/ru_sharegpt_cleaned",
    "dataset:IlyaGusev/ru_turbo_saiga",
    "dataset:lksy/ru_instruct_gpt4",
    "dataset:wikimedia/wikipedia",
    "dataset:uonlp/CulturaX",
    "dataset:allenai/c4",
    "base_model:NickupAI/LFM2.5-1.2B-Saiga-It-v2",
    "base_model:quantized:NickupAI/LFM2.5-1.2B-Saiga-It-v2",
    "license:other",
    "endpoints_compatible",
    "region:us",
    "imatrix",
    "conversational"
  ],
  "likes": 2,
  "downloads": 172,
  "gated": false,
  "private": false,
  "last_modified": "2026-03-12T09:23:08.000Z",
  "created_at": "2026-03-12T08:42:50.000Z",
  "pipeline_tag": "",
  "library_name": "transformers"
}

{
  "_id": "69b27c8aeee175d442f1bef7",
  "id": "mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF",
  "modelId": "mradermacher/LFM2.5-1.2B-Saiga-It-v2-i1-GGUF",
  "sha": "0092eee6358d25bcd7a7436a6761fee2f1df645a",
  "createdAt": "2026-03-12T08:42:50.000Z",
  "lastModified": "2026-03-12T09:23:08.000Z",
  "author": "mradermacher",
  "downloads": 172,
  "likes": 2,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "transformers",
  "siblings_count": 27
}