# AudioCraft (MusicGen)

> AudioCraft is Meta's PyTorch library with MusicGen (300M to 3.3B parameters), MAGNeT, JASCO and AudioGen. MusicGen makes instrumental music from a text description, and the melody models can follow a reference tune. It cannot sing: the model card says it does not generate realistic vocals, so it does not replace Suno's songs. Code is MIT, but all weights are CC BY-NC 4.0, so no commercial use. It needs a GPU (16 GB recommended) and pins Python 3.9 with PyTorch 2.1.

- Repository: https://github.com/facebookresearch/audiocraft
- Official docs: https://github.com/facebookresearch/audiocraft/blob/main/docs/MUSICGEN.md
- Checked: 2026-10-16

## What it replaces
- Suno: Instrumental tracks (Instrumental mode) (partial). Instrumental clips only; no vocals, and the weights are non-commercial.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Yes, Python library on a local GPU | https://github.com/facebookresearch/audiocraft/blob/main/docs/MUSICGEN.md | 2026-10-15 |
| Code licence | Confirmed | MIT | https://github.com/facebookresearch/audiocraft#license | 2026-10-15 |
| Model weights licence | No | CC BY-NC 4.0 (non-commercial) | https://github.com/facebookresearch/audiocraft#license | 2026-10-15 |
| Commercial use | No | No, the weights are non-commercial | https://huggingface.co/facebook/musicgen-large | 2026-10-15 |
| Paid services needed? | Confirmed | None required | https://github.com/facebookresearch/audiocraft#installation | 2026-10-15 |
| Hardware | Confirmed | GPU required; 16 GB recommended, smaller GPUs only for short clips or the small model | https://github.com/facebookresearch/audiocraft/blob/main/docs/MUSICGEN.md | 2026-10-15 |
| Operating systems | Partial | README documents apt-based ffmpeg installation on Linux; the official MusicGen FAQ links a community Windows tutorial. Native macOS support is not confirmed in the reviewed README and MusicGen guide | https://github.com/facebookresearch/audiocraft/blob/main/docs/MUSICGEN.md | 2026-10-16 |
| Setup effort | Confirmed | Medium; pip install audiocraft with PyTorch 2.1 and ffmpeg | https://github.com/facebookresearch/audiocraft#installation | 2026-10-15 |

GitHub: 23720 stars, 2703 forks, last push 2026-03-03, archived: no.
---
Source: RepoQuarry (https://repoquarry.dev/projects/audiocraft). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.