Meta MusicGen (AudioCraft)
MusicGen is Meta's open text-to-music model, distributed in the AudioCraft research library alongside AudioGen (text-to-sound), MAGNeT, MusicGen Style (text-and-style conditioning), JASCO (chord/melody/drum conditioning), EnCodec and the AudioSeal watermarker.
- Weights
- Downloadable
- API
- No
- Commercial use
- The licence forbids commercial use
It generates instrumental music from text and, in the melody variants, from a reference melody. It remains the most widely forked open music model but is licensed for non-commercial use only.
Split licence: the code is MIT but the model weights are CC-BY-NC 4.0, which forbids commercial use.
Specification
- What it takes in
- text, audio
- What it makes
- audio
- Maximum duration
- not established
- Controls
- style-prompting, reference-audio, extend
- Weights
- Downloadable
- API
- No
- Pricing
- Free to download and self-host; no vendor-hosted service.
- Released or updated
- not established
Maximum duration is not established here. That means no page we could retrieve stated it, not that the model lacks one.
Strengths
- Widely available, heavily documented and forked
- Family covers music, sound effects, style conditioning and chord/melody conditioning
- MIT-licensed code makes research and tooling easy
- Includes AudioSeal watermarking
Weaknesses
- Weights are CC-BY-NC 4.0 - non-commercial only, which rules out most video production use
- Repository documents no maximum generation duration
- No vocals
- No first-party hosted API
- Quality now behind Suno, Lyria 3 and Stable Audio 3
Best for
- Research, prototyping and personal projects
- Non-commercial experimentation and fine-tuning
Where the sources disagree
Recorded rather than resolved. Picking the tidier number would hide the fact that the vendor's own pages do not agree.
- The repository README does not state release dates or maximum generation length for any model, so both are null.
What to understand before using it
Explainers naming the mechanisms this model's own documentation mentions.
See it from the vendor
Shotcraft hosts no sample output. Judging a model means watching its own showreel with your own eyes, and a clip re-hosted here would be a clip we chose.
Nearby in the directory
ACE-Step
An open-source music generation foundation model combining diffusion with a lightweight linear transformer, released May 2025 by ACE Studio and StepFu
Music and sound
AIVA
Generates instrumental music from style presets and influence uploads for video, game and content creators, exporting MP3 and MIDI on lower tiers and
Music and sound
Beatoven.ai maestro
Generates royalty-free background music and text-prompted sound effects for video, podcast, game and advertising work, downloadable as MP3 or WAV with
Music and sound