unsloth/falcon-h1-3b-instruct-gguf overview

{
  "metadata": {},
  "card_data": {
    "library_name": "transformers",
    "tags": [
      "falcon-h1",
      "unsloth"
    ],
    "license": "other",
    "license_name": "falcon-llm-license",
    "license_link": "https://falconllm.tii.ae/falcon-terms-and-conditions.html",
    "base_model": [
      "tiiuae/Falcon-H1-3B-Instruct"
    ],
    "inference": true,
    "frontmatter": {
      "library_name": "transformers",
      "tags": [
        "falcon-h1",
        "unsloth"
      ],
      "license": "other",
      "license_name": "falcon-llm-license",
      "license_link": "https://falconllm.tii.ae/falcon-terms-and-conditions.html",
      "base_model": [
        "tiiuae/Falcon-H1-3B-Instruct"
      ],
      "inference": "true"
    },
    "hero_image_url": "https://github.com/unslothai/unsloth/raw/main/images/unsloth%20new%20logo.png",
    "summary": "0. TL;DR 1. Model Details 2. Training Details 3. Usage 4. Evaluation 5. Citation # TL;DR # Model Details",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlibrary_name: transformers\ntags:\n- falcon-h1\n- unsloth\nlicense: other\nlicense_name: falcon-llm-license\nlicense_link: https://falconllm.tii.ae/falcon-terms-and-conditions.html\nbase_model:\n- tiiuae/Falcon-H1-3B-Instruct\ninference: true\n---\n> [!NOTE]\n>  Includes our **chat template fixes**! <br> For `llama.cpp`, use `--jinja`\n>\n\n<div>\n<p style=\"margin-top: 0;margin-bottom: 0;\">\n    <em><a href=\"https://docs.unsloth.ai/basics/unsloth-dynamic-v2.0-gguf\">Unsloth Dynamic 2.0</a> achieves superior accuracy & outperforms other leading quants.</em>\n  </p>\n  <div style=\"display: flex; gap: 5px; align-items: center; \">\n    <a href=\"https://github.com/unslothai/unsloth/\">\n      <img src=\"https://github.com/unslothai/unsloth/raw/main/images/unsloth%20new%20logo.png\" width=\"133\">\n    </a>\n    <a href=\"https://discord.gg/unsloth\">\n      <img src=\"https://github.com/unslothai/unsloth/raw/main/images/Discord%20button.png\" width=\"173\">\n    </a>\n    <a href=\"https://docs.unsloth.ai/\">\n      <img src=\"https://raw.githubusercontent.com/unslothai/unsloth/refs/heads/main/images/documentation%20green%20button.png\" width=\"143\">\n    </a>\n  </div>\n</div>\n\n\n<img src=\"https://huggingface.co/datasets/tiiuae/documentation-images/resolve/main/falcon_mamba/falcon-h1-logo.png\" alt=\"drawing\" width=\"800\"/>\n\n#  Table of Contents\n\n0. [TL;DR](#TL;DR)\n1. [Model Details](#model-details)\n2. [Training Details](#training-details)\n3. [Usage](#usage)\n4. [Evaluation](#evaluation)\n5. [Citation](#citation)\n\n# TL;DR\n\n# Model Details\n\n## Model Description\n\n- **Developed by:** [https://www.tii.ae](https://www.tii.ae)\n- **Model type:** Causal decoder-only\n- **Architecture:** Hybrid Transformers + Mamba architecture\n- **Language(s) (NLP):** English, Multilingual\n- **License:** Falcon-LLM License\n\n# Training details\n\nFor more details about the training protocol of this model, please refer to the [Falcon-H1 technical blogpost](https://falcon-lm.github.io/blog/falcon-h1/).\n\n# Usage\n\nCurrently to use this model you can either rely on Hugging Face `transformers`, `vLLM` or `llama.cpp` library.\n\n## Inference\n\nMake sure to install the latest version of `transformers` or `vllm`, eventually install these packages from source:\n\n```bash\npip install git+https://github.com/huggingface/transformers.git\n```\n\nFor vLLM, make sure to install `vllm>=0.9.0`:\n\n```bash\npip install \"vllm>=0.9.0\"\n```\n\n### 🤗 transformers\n\nRefer to the snippet below to run H1 models using 🤗 transformers:\n\n```python\nimport torch\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel_id = \"tiiuae/Falcon-H1-1B-Base\"\n\nmodel = AutoModelForCausalLM.from_pretrained(\n  model_id,\n  torch_dtype=torch.bfloat16,\n  device_map=\"auto\"\n)\n\n# Perform text generation\n```\n\n### vLLM\n\nFor vLLM, simply start a server by executing the command below:\n\n```\n# pip install vllm\nvllm serve tiiuae/Falcon-H1-1B-Instruct --tensor-parallel-size 2 --data-parallel-size 1\n```\n\n### `llama.cpp`\n\nYou can find all GGUF files under [our official collection](https://huggingface.co/collections/tiiuae/falcon-h1-6819f2795bc406da60fab8df)\n\n# Evaluation\n\nFalcon-H1 series perform very well on a variety of tasks, including reasoning tasks. \n\n| Tasks | Falcon-H1-3B | Qwen3-4B | Qwen2.5-3B | Gemma3-4B | Llama3.2-3B | Falcon3-3B |\n| --- | --- | --- | --- | --- | --- | --- |\n| **General**  | | | | | |\n| BBH | **53.69** | 51.07 | 46.55 | 50.01 | 41.47 | 45.02 |\n| ARC-C | **49.57** | 37.71 | 43.77 | 44.88 | 44.88 | 48.21 |\n| TruthfulQA | 53.19 | 51.75 | **58.11** | 51.68 | 50.27 | 50.06 |\n| HellaSwag | **69.85** | 55.31 | 64.21 | 47.68 | 63.74 | 64.24 |\n| MMLU | **68.3** | 67.01 | 65.09 | 59.53 | 61.74 | 56.76 |\n| **Math**  | | | | | |\n| GSM8k | **84.76** | 80.44 | 57.54 | 77.41 | 77.26 | 74.68 |\n| MATH-500 | 74.2 | **85.0** | 64.2 | 76.4 | 41.2 | 54.2 |\n| AMC-23 | 55.63 | **66.88** | 39.84 | 48.12 | 22.66 | 29.69 |\n| AIME-24 | 11.88 | **22.29** | 6.25 | 6.67 | 11.67 | 3.96 |\n| AIME-25 | 13.33 | **18.96** | 3.96 | 13.33 | 0.21 | 2.29 |\n| **Science**  | | | | | |\n| GPQA | **33.89** | 28.02 | 28.69 | 29.19 | 28.94 | 28.69 |\n| GPQA_Diamond | 38.72 | **40.74** | 35.69 | 28.62 | 29.97 | 29.29 |\n| MMLU-Pro | **43.69** | 29.75 | 32.76 | 29.71 | 27.44 | 29.71 |\n| MMLU-stem | **69.93** | 67.46 | 59.78 | 52.17 | 51.92 | 56.11 |\n| **Code**  | | | | | |\n| HumanEval | 76.83 | **84.15** | 73.78 | 67.07 | 54.27 | 52.44 |\n| HumanEval+ | 70.73 | **76.83** | 68.29 | 61.59 | 50.0 | 45.73 |\n| MBPP | **79.63** | 68.78 | 72.75 | 77.78 | 62.17 | 61.9 |\n| MBPP+ | **67.46** | 59.79 | 60.85 | 66.93 | 50.53 | 55.29 |\n| LiveCodeBench | 26.81 | **39.92** | 11.74 | 21.14 | 2.74 | 3.13 |\n| CRUXEval | 56.25 | **69.63** | 43.26 | 52.13 | 17.75 | 44.38 |\n| **Instruction Following**  | | | | | |\n| IFEval | **85.05** | 84.01 | 64.26 | 77.01 | 74.0 | 69.1 |\n| Alpaca-Eval | 31.09 | 36.51 | 17.37 | **39.64** | 19.69 | 14.82 |\n| MTBench | **8.72** | 8.45 | 7.79 | 8.24 | 7.96 | 7.79 |\n| LiveBench | 36.86 | **51.34** | 27.32 | 36.7 | 26.37 | 26.01 |\n\nYou can check more in detail on our [our release blogpost](https://falcon-lm.github.io/blog/falcon-h1/), detailed benchmarks.\n\n# Useful links\n\n- View [our release blogpost](https://falcon-lm.github.io/blog/falcon-h1/).\n- Feel free to join [our discord server](https://discord.gg/trwMYP9PYm) if you have any questions or to interact with our researchers and developers.\n\n# Citation\n\nIf the Falcon-H1 family of models were helpful to your work, feel free to give us a cite.\n\n```\n@misc{tiifalconh1,\n    title = {Falcon-H1: A Family of Hybrid-Head Language Models Redefining Efficiency and Performance},\n    url = {https://falcon-lm.github.io/blog/falcon-h1},\n    author = {Falcon-LLM Team},\n    month = {May},\n    year = {2025}\n}\n```",
    "related_quantizations": []
  },
  "tags": [
    "transformers",
    "gguf",
    "falcon-h1",
    "unsloth",
    "base_model:tiiuae/Falcon-H1-3B-Instruct",
    "base_model:quantized:tiiuae/Falcon-H1-3B-Instruct",
    "license:other",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 3,
  "downloads": 227,
  "gated": false,
  "private": false,
  "last_modified": "2025-07-10T08:46:23.000Z",
  "created_at": "2025-07-10T07:55:17.000Z",
  "pipeline_tag": "",
  "library_name": "transformers"
}

{
  "_id": "686f71e58d0d1c665d18b8d4",
  "id": "unsloth/Falcon-H1-3B-Instruct-GGUF",
  "modelId": "unsloth/Falcon-H1-3B-Instruct-GGUF",
  "sha": "02cb641c60514333e2d46288a55a27b83905f0eb",
  "createdAt": "2025-07-10T07:55:17.000Z",
  "lastModified": "2025-07-10T08:46:23.000Z",
  "author": "unsloth",
  "downloads": 227,
  "likes": 3,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "transformers",
  "siblings_count": 29
}