{
  "id": "demucs",
  "name": "Demucs",
  "category": "music_audio",
  "description": "Separate a mixed music recording into vocals, drums, bass and other accompaniment with a locally runnable source-separation model.",
  "useCases": [
    "Isolate parts from your own recordings for practice and analysis",
    "Prepare vocal and instrumental stems for permitted remix or editing work"
  ],
  "repoUrl": "https://github.com/adefossez/demucs",
  "originalRepoUrl": "https://github.com/facebookresearch/demucs",
  "docsUrl": "https://github.com/adefossez/demucs#for-musicians",
  "demo": {
    "url": "https://ai.honu.io/papers/htdemucs/index.html",
    "label": "Compare the creators’ separated stems",
    "evidenceType": "creator_demo",
    "note": "Paper authors’ comparison samples, linked from the repository. Includes models beyond the standard downloadable package; no separation was run for this directory."
  },
  "license": {
    "code": "MIT",
    "weights": "Not separately specified in the inspected official HTDemucs weights repository",
    "caveat": "Both code repositories contain MIT LICENSE files. The current official HTDemucs model card and file list omit a weights license; do not automatically extend the code license to every checkpoint. Verify weights terms before deployment. Input music rights are separate.",
    "url": "https://github.com/adefossez/demucs/blob/main/LICENSE",
    "weightsUrl": "https://huggingface.co/adefossez/HTDemucs"
  },
  "pricing": {
    "summary": "No code license fee for local separation; you supply the compute.",
    "caveat": "CPU mode is available. The README estimates roughly 7 GB GPU memory for default arguments, with lower-memory settings available; actual requirements and time depend on the model and track."
  },
  "platforms": [
    "Linux",
    "macOS",
    "Windows",
    "Command line",
    "Python API"
  ],
  "setupDifficulty": "moderate",
  "setupRequirements": [
    "Current official continuation requires Python 3.10+ and PyTorch 2.1+",
    "Use the documented uv/uvx or pip workflow; model checkpoints download separately",
    "FFmpeg is recommended and needed for FLAC output or otherwise unsupported input formats",
    "Intel Macs have an upstream Python/PyTorch version caveat; consult current instructions"
  ],
  "maintenance": {
    "status": "limited_maintenance",
    "note": "Meta’s original repository was archived on 2025-01-01. Alexandre Défossez’s official continuation released v4.1.0 on 2026-07-11, but still warns of slow replies and no new features.",
    "successorUrl": "https://github.com/adefossez/demucs",
    "releaseNotesUrl": "https://github.com/adefossez/demucs/blob/main/docs/release.md"
  },
  "limitations": [
    "Separation can leave bleed or artifacts; experimental piano separation is specifically caveated",
    "Two-stem mode still performs full separation internally",
    "Some sparse paper models are not supplied in the standard release",
    "Archived original, limited-maintenance continuation and unclear standalone weights terms require extra diligence"
  ],
  "evidenceCheckedDate": "2026-10-14",
  "testStatus": "not_tested",
  "sources": [
    {
      "title": "Archived Meta original and official continuation pointer",
      "url": "https://github.com/facebookresearch/demucs",
      "evidenceCheckedDate": "2026-10-14"
    },
    {
      "title": "Official continuation: maintenance, setup and usage",
      "url": "https://github.com/adefossez/demucs",
      "evidenceCheckedDate": "2026-10-14"
    },
    {
      "title": "Actual original code LICENSE: MIT",
      "url": "https://github.com/facebookresearch/demucs/blob/main/LICENSE",
      "evidenceCheckedDate": "2026-10-14"
    },
    {
      "title": "Actual continuation code LICENSE: MIT",
      "url": "https://github.com/adefossez/demucs/blob/main/LICENSE",
      "evidenceCheckedDate": "2026-10-14"
    },
    {
      "title": "Release notes: v4.1.0, July 11, 2026",
      "url": "https://github.com/adefossez/demucs/blob/main/docs/release.md",
      "evidenceCheckedDate": "2026-10-14"
    },
    {
      "title": "Official model card without separate license declaration",
      "url": "https://huggingface.co/adefossez/HTDemucs",
      "evidenceCheckedDate": "2026-10-14"
    },
    {
      "title": "Official model files and visible license-removal commit",
      "url": "https://huggingface.co/adefossez/HTDemucs/tree/main",
      "evidenceCheckedDate": "2026-10-14"
    },
    {
      "title": "Creators’ source-separation examples",
      "url": "https://ai.honu.io/papers/htdemucs/index.html",
      "evidenceCheckedDate": "2026-10-14"
    }
  ],
  "categoryLabel": "Music & Audio",
  "tags": [
    "stem separation",
    "vocals",
    "music editing",
    "local",
    "limited maintenance"
  ],
  "card": {
    "cost": "Local compute; check weights terms",
    "setup": "Python + audio dependencies"
  },
  "setup": {
    "summary": "Python + audio dependencies",
    "requirements": [
      "Current official continuation requires Python 3.10+ and PyTorch 2.1+",
      "Use the documented uv/uvx or pip workflow; model checkpoints download separately",
      "FFmpeg is recommended and needed for FLAC output or otherwise unsupported input formats",
      "Intel Macs have an upstream Python/PyTorch version caveat; consult current instructions"
    ]
  },
  "evidence": {
    "status": "documentation_reviewed",
    "demoType": "creator_demo",
    "independentlyTested": false,
    "note": "Official source documentation reviewed; no repository installed or executed."
  },
  "releaseDate": null,
  "releaseDateNote": "Not researched; the evidence check date does not indicate a recent release."
}