Skip to content
← Models
Open weightsMeta/10 trustMetrics-checked

MusicGen Stereo

Open stereo music-generation models from Meta's AudioCraft family.

Visit official page →

About

MusicGen Stereo is the stereo-output branch of Meta's MusicGen text-to-music models, released as part of the open-source AudioCraft library. While the base MusicGen models (300M to 3.3B) generate mono audio, the stereo variants (facebook/musicgen-stereo-small through musicgen-stereo-large) produce true stereo output, with melody-conditioned variants that can reimagine a reference melody in new styles.

Generation caps at around 30 seconds per call, chainable for longer tracks, and the models run in 12GB VRAM at fp16 via the audiocraft pip package or Hugging Face Transformers. Important license caveat: MusicGen is released under CC BY-NC 4.0, which prohibits commercial use, unlike MIT-licensed siblings in the family.

Specs

Context window

Pricing

Free quota

API access

Weights

Open weights

Released

November 1, 2023

Benchmarks

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

music-generationbackground-musicaudio-prototyping

Trust score — 7/10

/10 trustMetrics-checked

Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 84; Trending score 72/100 from researcher assessment.

Sources