liquidai/lfm2-audio-1.5b-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "other",
    "license_name": "lfm1.0",
    "license_link": "LICENSE",
    "tags": [
      "audio",
      "liquid",
      "lfm2",
      "edge",
      "llama.cpp",
      "gguf"
    ],
    "base_model": [
      "LiquidAI/LFM2-Audio-1.5B"
    ],
    "frontmatter": {
      "license": "other",
      "license_name": "lfm1.0",
      "license_link": "LICENSE",
      "tags": [
        "audio",
        "liquid",
        "lfm2",
        "edge",
        "llama.cpp",
        "gguf"
      ],
      "base_model": [
        "LiquidAI/LFM2-Audio-1.5B"
      ]
    },
    "hero_image_url": "https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/2b08LKpev0DNEk6DlnWkY.png",
    "summary": "This example demonstrates the **LFM2-Audio-1.5B** audio model. Link to HF: LiquidAI/LFM2-Audio-1.5B. The model supports following modes",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: other\nlicense_name: lfm1.0\nlicense_link: LICENSE\ntags:\n- audio\n- liquid\n- lfm2\n- edge\n- llama.cpp\n- gguf\nbase_model:\n- LiquidAI/LFM2-Audio-1.5B\n---\n\n<center>\n<div style=\"text-align: center;\">\n  <img \n    src=\"https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/2b08LKpev0DNEk6DlnWkY.png\" \n    alt=\"Liquid AI\"\n    style=\"width: 100%; max-width: 100%; height: auto; display: inline-block; margin-bottom: 0.5em; margin-top: 0.5em;\"\n  />\n</div>\n<div style=\"display: flex; justify-content: center; gap: 0.5em;\">\n  <a href=\"https://playground.liquid.ai/chat\">\n<a href=\"https://playground.liquid.ai/\"><strong>Try LFM</strong></a> • <a href=\"https://docs.liquid.ai/lfm\"><strong>Documentation</strong></a> • <a href=\"https://leap.liquid.ai/\"><strong>LEAP</strong></a></a>\n</div>\n</center>\n\n# LFM2-Audio-1.5B-GGUF\n\nThis example demonstrates the **LFM2-Audio-1.5B** audio model.\n\nLink to HF: [LiquidAI/LFM2-Audio-1.5B](https://huggingface.co/LiquidAI/LFM2-Audio-1.5B).\n\nThe model supports following modes\n- ASR:\n  - input `audio.wav`, output `text`\n- TTS:\n  - input `text`, output `audio.wav`\n- interleaved:\n  - input `text` or `audio.wav`, output `text` and `audio.wav`\n\n## GGUFS\n\nThere are total 3 GGUFs for this model.\n\nSet `$CKPT` to path to the path containing downloaded GGUFs.\nSet `$INPUT_WAV` to path to input wav file.\n\n```console\nexport CKPT=/data/playground/checkpoints/LFM2-Audio-1.5B-GGUF\nexport INPUT_WAV=/tmp/input.wav\nexport OUTPUT_WAV=/tmp/output.wav\n```\n\n```console\n(cd $CKPT && ls *.gguf)\naudiodecoder-LFM2-Audio-1.5B-Q8_0.gguf  LFM2-Audio-1.5B-Q8_0.gguf  mmproj-audioencoder-LFM2-Audio-1.5B-Q8_0.gguf\n```\n\nOptionally, float16 ggufs can be downloaded and used by replacing `Q8_0` with `F16`.\n\n## Binaries\n\n`runners` folder contain runners for andoird-arm64, macos-arm64, ubuntu-arm64, and ubuntu-x64.\n```console\nrunners\n├── android-arm64\n│   └── lfm2-audio-android-arm64.zip\n├── macos-arm64\n│   └── lfm2-audio-macos-arm64.zip\n├── ubuntu-arm64\n│   └── lfm2-audio-ubuntu-arm64.zip\n└── ubuntu-x64\n    └── lfm2-audio-ubuntu-x64.zip\n```\n\nEach package contains `llama-lfm2-audio` and `llama-mtmd-cli` binaries.\n\n## Run using `llama-lfm2-audio`\n\nThere are 3 supported modes\n- ASR\n- TTS\n- interleaved\n\nThe mode is defined by system prompt. There are limitations on system prompt and binary will check for them and raise an error if needed.\n\n### ASR\n\nASR requires `-sys \"Perform ASR.\"` and `--audio audio.wav` for input. It will print text to console\n\n```console\nlfm2-audio-<platform>/llama-lfm2-audio -m $CKPT/LFM2-Audio-1.5B-Q8_0.gguf --mmproj $CKPT/mmproj-audioencoder-LFM2-Audio-1.5B-Q8_0.gguf -mv $CKPT/audiodecoder-LFM2-Audio-1.5B-Q8_0.gguf -sys \"Perform ASR.\" --audio $INPUT_WAV\n```\n\n### TTS\n\nTTS requires `-sys \"Perform TTS.\"`, `-p \"What is this obsession people have with books?\"` for input, and `--output output.wav` for output. It will save audio to `output.wav`.\n```console\nlfm2-audio-<platform>/llama-lfm2-audio -m $CKPT/LFM2-Audio-1.5B-Q8_0.gguf --mmproj $CKPT/mmproj-audioencoder-LFM2-Audio-1.5B-Q8_0.gguf -mv $CKPT/audiodecoder-LFM2-Audio-1.5B-Q8_0.gguf -sys \"Perform TTS.\" -p \"What is this obsession people have with books?\" --output $OUTPUT_WAV\n```\n\n### Interleaved\n\nInterleaved produces both, text and audio as output, and can consume text or audio as input.\n\n```console\nlfm2-audio-<platform>/llama-lfm2-audio -m $CKPT/LFM2-Audio-1.5B-Q8_0.gguf --mmproj $CKPT/mmproj-audioencoder-LFM2-Audio-1.5B-Q8_0.gguf -mv $CKPT/audiodecoder-LFM2-Audio-1.5B-Q8_0.gguf -sys \"Respond with interleaved text and audio.\" --audio $INPUT_WAV --output $OUTPUT_WAV\n```\n\n## Run ASR using `llama-mtmd-cli`\n\nBuild `llama-mtmd-cli` following the standard build procedure.\n\n```console\nlfm2-audio-<platform>/llama-mtmd-cli -m $CKPT/LFM2-Audio-1.5B-Q8_0.gguf --mmproj $CKPT/mmproj-audioencoder-LFM2-Audio-1.5B-Q8_0.gguf -p \"<__media__>\" -sys \"Perform ASR.\" --audio $INPUT_WAV\n```\n\n### Debug\n\nFor reproducible results set `--temp 0`.\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "audio",
    "liquid",
    "lfm2",
    "edge",
    "llama.cpp",
    "base_model:LiquidAI/LFM2-Audio-1.5B",
    "base_model:quantized:LiquidAI/LFM2-Audio-1.5B",
    "license:other",
    "region:us"
  ],
  "likes": 29,
  "downloads": 766,
  "gated": false,
  "private": false,
  "last_modified": "2025-12-05T17:50:52.000Z",
  "created_at": "2025-09-30T12:41:36.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "68dbd000fc017b004167352b",
  "id": "LiquidAI/LFM2-Audio-1.5B-GGUF",
  "modelId": "LiquidAI/LFM2-Audio-1.5B-GGUF",
  "sha": "5a54beabc49a8abcd158a9ea86516fa4ae82dfc9",
  "createdAt": "2025-09-30T12:41:36.000Z",
  "lastModified": "2025-12-05T17:50:52.000Z",
  "author": "LiquidAI",
  "downloads": 766,
  "likes": 29,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 15
}