enlistedghost/pixtral-12b-ollama-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "language": [
      "en",
      "ru"
    ],
    "base_model": [
      "mistral-community/pixtral-12b"
    ],
    "pipeline_tag": "image-text-to-text",
    "new_version": "EnlistedGhost/Pixtral-12B-Ollama-GGUF",
    "datasets": [
      "mistralai/MM-MT-Bench"
    ],
    "tags": [
      "Pixtral,",
      "12B,",
      "Vision,",
      "Conversational,",
      "Ollama,",
      "ggml,",
      "gguf,",
      "Image-Text-to-Text,",
      "Multimodal,"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "language": [
        "en",
        "ru"
      ],
      "base_model": [
        "mistral-community/pixtral-12b"
      ],
      "pipeline_tag": "image-text-to-text",
      "new_version": "EnlistedGhost/Pixtral-12B-Ollama-GGUF",
      "datasets": [
        "mistralai/MM-MT-Bench"
      ],
      "tags": [
        "Pixtral,",
        "12B,",
        "Vision,",
        "Conversational,",
        "Ollama,",
        "ggml,",
        "gguf,",
        "Image-Text-to-Text,",
        "Multimodal,"
      ]
    },
    "hero_image_url": "",
    "summary": "# Model - Pixtral 12B (GGUF) (Ollama Patched) **Description:**  This is an Ollama patched version of Pixtral-12B (GGUF) with working projector (mmproj) files.  No modifications, edits, or configuration is required to use this model with Ollama, it works natively!  Both Vision and Text work with Ollama. (^.^) **IMPORTANT NOTICE as of **(GMT-8) 01:04 November 19th, 2025**:**  After significant amounts of self-testing of the supplied Quantized GGUF files made available within this release  I have been pleasantly surprised to find that so far ALL of the files have the AI performing quite admirably despite  Huggingface's systems here only merging in the Q8_0 (8-bit quantized) Vision Projector - with no automated option to  choose a higher quality beyond assuming that you already know how to use Ollama's \"create\" command and so on. **I will be updating this release with both:** Multi-Modal model that you may speak/inference with utilizing Ollama (or llama.cpp as well). simply download the small script file and then merely run (click on) the script which will interactively guide you  through a user-friendly setup process - including merging the higher quality Vision Projector options.",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\nlanguage:\n- en\n- ru\nbase_model:\n- mistral-community/pixtral-12b\npipeline_tag: image-text-to-text\nnew_version: EnlistedGhost/Pixtral-12B-Ollama-GGUF\ndatasets:\n- mistralai/MM-MT-Bench\ntags:\n- Pixtral,\n- 12B,\n- Vision,\n- Conversational,\n- Ollama,\n- ggml,\n- gguf,\n- Image-Text-to-Text,\n- Multimodal,\n---\n# Model - Pixtral 12B (GGUF) (Ollama Patched)\n**Description:** <br />\nThis is an Ollama patched version of Pixtral-12B (GGUF) with working projector (mmproj) files. <br />\nNo modifications, edits, or configuration is required to use this model with Ollama, it works natively! <br />\nBoth Vision and Text work with Ollama. (^.^)\n\n**IMPORTANT NOTICE as of **(GMT-8) 01:04 November 19th, 2025**:** <br /> \nAfter significant amounts of self-testing of the supplied Quantized GGUF files made available within this release <b />\nI have been pleasantly surprised to find that so far ALL of the files have the AI performing quite admirably despite <b />\nHuggingface's systems here only merging in the Q8_0 (8-bit quantized) Vision Projector - with no automated option to <b />\nchoose a higher quality beyond assuming that you already know how to use Ollama's \"create\" command and so on.\n\n**I will be updating this release with both:**\n- Instructions within this model page and linked documentation on how to easily \"create\" a customized/tailored <br />\nMulti-Modal model that you may speak/inference with utilizing Ollama (or llama.cpp as well).\n- Automated scripts via bash (Linux users) and command-line (Windows users) options that will allow anyone to <br />\nsimply download the small script file and then merely run (click on) the script which will interactively guide you <b />\nthrough a user-friendly setup process - including merging the higher quality Vision Projector options.\n\n## In addition to the above mentioned updates:\nI am currently releasing GGUF Ollama-ready (Natively Ollama compatible) alternate versions of <b />\nPixtral-12B. **In case some of you are not aware of this fact:** there are currently two other Officially released versions <b />\nof Pixtral-12B. **These are supplied directly from MistralAI themselves** and include the Pixtral-12B-2409 official release and <b />\nthe the Pixtral-12B-Base-2409 models that, unlike the release that these files were converted and quantized from, ARE NOT a community release. <br />\nThese two other versions differ in quality and capability, although slightly, from the originating \"mistral-community\" release <b />\nthat this GGUF release (and MOST other releases or derivatives) is based upon. This is because MistralAI's two releases with the \"2409\" label <b />\nare not user-friendly to setup or use. Thankfully I have humbly managed to have them successfully loaded, converted to GGUF and now <b />\nam in the process of creating quality Quantized GGUF versions of specifically for use with Ollama/Llama.cpp. \n\n**Be sure to check back occasionally as I expect to make these more rarely seen Pixtral-12B releases from MistralAI made available within the next couple days!**\n*(Rarely seen - to clarify what I meant was that they are extremely uncommong to find GGUF versions of, and not many versions even that are not GGUF)*\n\nHappy Inferencing! <br />\n -- Jon Z (EnlistedGhost)\n\n---------------------------------------------\n\n### Model Updates (As of: November 19th, 2025)\nRecently finished updates:\n- Partially Updated: ModelCard <br /> \n(this page)\n\nCurrently in-progress updates:\n- Update: ModelCard <br /> \n(this page needs more setup guides and information...)\n\n## How to run this Model using Ollama\nYou can run this model by using the \"ollama run\" command.<br />\nSimply copy & paste one of the commands from the list below into<br /> \nyour console, terminal or power-shell window.\n| Quant Type | File Size | Command |\n|:-----------|:----------|:--------|\n| Q2_K_S | 4.62 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q2_K_S |\n| Q2_K  | 4.79 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q2_K |\n| IQ2_M | 4.44 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:IQ2_M |\n| Q2_K_M | 4.95 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q2_K_M |\n| Q2_K_L  | 5.58 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q2_K_L |\n| IQ3_S | 5.56 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:IQ3_S |\n| Q3_K_S | 5.53 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q3_K_S |\n| IQ3_M | 5.72 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:IQ3_M |\n| Q3_K_M | 6.08 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q3_K_M |\n| Q3_K_L | 6.56 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q3_K_L |\n| Q3_K_XL | 6.72 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q3_K_XL |\n| IQ4_XS | 6.8 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:IQ4_XS |\n| Q4_K_S | 7.12 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q4_K_S |\n| Q4_K_M  | 7.48 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q4_K_M |\n| Q4_K_XL |8.27 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q4_K_XL |\n| Q5_K_S | 8.52 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q5_K_S |\n| Q5_K_M  | 8.73 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q5_K_M |\n| Q5_K_XL | 9.52 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q5_K_XL |\n| Q6_K | 10.1 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q6_K |\n| Q6_K_M | 10.2 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q6_K_M |\n| Q6_K_L | 10.8 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q6_K_L |\n| Q8_0 | 13.7 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:Q8_0 |\n| F16 | 24.5 GB | ollama run hf.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF:F16 \n\n-------------------------------------------------------------------------------\n*mmproj (Vision Projector) Files*\n| Quant Type | File Size | Command |\n|:-----------|:----------|:--------|\n| Q8_0 | 465 MB | [[mmproj Vision Pixtral-12B rojector:Q8_0](https://huggingface.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF/resolve/main/mmproj-Pixtral-12B-Q8_0.gguf?download=true)]|\n| F16 | 870 MB | [[mmproj Vision Pixtral-12B rojector:F16](https://huggingface.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF/resolve/main/mmproj-Pixtral-12B-F16.gguf?download=true)]|\n| F32 | 1.74 GB | [[mmproj Vision Pixtral-12B rojector:F32](https://huggingface.co/EnlistedGhost/Pixtral-12B-Ollama-GGUF/resolve/main/mmproj-Pixtral-12B-F32.gguf?download=true)]|\n\n### Intended Use\nSame as original:\n- [[mistralai](https://huggingface.co/mistralai/Pixtral-12B-2409)]\n\n\n### Out-of-Scope Use\nSame as original:\n- [[mistralai](https://huggingface.co/mistralai/Pixtral-12B-2409)]\n\n\n## Bias, Risks, and Limitations\nSame as original:\n- [[mistralai](https://huggingface.co/mistralai/Pixtral-12B-2409)]\n\n\n## Training Details\nTraining sets and data are from:\n- [[mistralai](https://huggingface.co/mistralai/Pixtral-12B-2409)]\n- [[mistralai-community](https://huggingface.co/mistral-community/pixtral-12b)]\n<br />(This is a direct off-shoot/decendant of the above mentioned models)\n\n\n## Evaluation\n- This model has NOT been evaluated in any form, scope or type of method. \n- **!!! USE AT YOUR OWN RISK !!!**\n- **!!! NO WARRANTY IS PROVIDED OF ANY KIND !!!**\n\n## Citation (Original Paper)\n[[MistalAI Pixtral-12B Original Paper](https://huggingface.co/papers/2410.07073)]\n\n\n## Detailed Release Information\n- **Originally Developed by:** [[mistralai](https://huggingface.co/mistralai/Pixtral-12B-2409)]\n- **Further Developed by:** [[mistralai-community](https://huggingface.co/mistral-community/pixtral-12b)]\n- **MMPROJ (Vision) Quantized by:** [[EnlistedGhost](https://huggingface.co/EnlistedGhost)]\n- **Model Quantized for GGUF by:** [[EnlistedGhost](https://huggingface.co/EnlistedGhost)]\n- **Modified for Ollama by:** [[EnlistedGhost](https://huggingface.co/EnlistedGhost)]\n- **Released on Huggingface by:** [[EnlistedGhost](https://huggingface.co/EnlistedGhost)]\n- **Model type & format:** [Quantized/GGUF]\n- **License type:** [Apache-2.0]\n\n\n## Attributions (Credits)\nA big thank-you is extended to the below credited sources!<br />\nThese contributions are what made this release possible!\n\n*Important Notice: **This is NOT a copy/paste release**, <br />\nI have created unique Quantized files that were then altered further <br />\nto properly work with Ollama software. <br />\nThis resulted in the first publicly available <br />\nPixtral-12B model that natively runs on Ollama.*\n\n- <s> [[ggml-org](https://huggingface.co/ggml-org/pixtral-12b-GGUF)] </s> (Resources no longer utilized)\n- <s> [[mrmadermacher](https://huggingface.co/mradermacher/pixtral-12b-GGUF)] </s> (Resources no longer utilized)\n-  [[bartowski](https://huggingface.co/bartowski/mistral-community_pixtral-12b-GGUF)] (Still using low iMatrix Quant files: IQ2_M and IQ2_S)\n- [[Ollama developers](https://github.com/ollama/ollama/issues/6748#issuecomment-3449009817)]\n\n## Model Card Authors and Contact\n[[EnlistedGhost](https://huggingface.co/EnlistedGhost)]",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "Pixtral,",
    "12B,",
    "Vision,",
    "Conversational,",
    "Ollama,",
    "ggml,",
    "gguf,",
    "Image-Text-to-Text,",
    "Multimodal,",
    "image-text-to-text",
    "en",
    "ru",
    "dataset:mistralai/MM-MT-Bench",
    "arxiv:2410.07073",
    "base_model:mistral-community/pixtral-12b",
    "base_model:quantized:mistral-community/pixtral-12b",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 2,
  "downloads": 2033,
  "gated": false,
  "private": false,
  "last_modified": "2025-11-20T08:37:31.000Z",
  "created_at": "2025-10-19T20:42:40.000Z",
  "pipeline_tag": "image-text-to-text",
  "library_name": ""
}

{
  "_id": "68f54d406ab141b2a3294b5c",
  "id": "EnlistedGhost/Pixtral-12B-Ollama-GGUF",
  "modelId": "EnlistedGhost/Pixtral-12B-Ollama-GGUF",
  "sha": "d5660ebd9400c005e76116f71f2518d666d42cc3",
  "createdAt": "2025-10-19T20:42:40.000Z",
  "lastModified": "2025-11-20T08:37:31.000Z",
  "author": "EnlistedGhost",
  "downloads": 2033,
  "likes": 2,
  "gated": false,
  "private": false,
  "pipeline_tag": "image-text-to-text",
  "library_name": "",
  "siblings_count": 46
}