m-a-p · Open Music Weights
YuE2
Moved from Week 37 2026, where it was covered from the project page. The weights are now downloadable: lyrics and a style prompt become a full song with vocals and accompaniment through an AR-NAR mixture-of-transformers backbone writing score and semantic tokens, with flow matching for the acoustic latents, plus cover-song generation via melody transcription. Runs locally on a 24GB GPU in BF16 without quantisation, licensed CC BY-NC 4.0 so non-commercial only. On WildSongBench it reports 6.9632 best-of-8 against Suno v5 at 6.8721 and v6 at 6.5562. One caution: the repo is named YuE2-3B but its own tensor spec says 4B, so treat the parameter count as unsettled.
Open the source ↗Similar Audio & Speech releases
- Suno v6
Suno · Suno V6 · Week 37, 2026
- Happy Shrimp
Happy Shrimp · Music Generation · Week 34, 2026
- MiniMax Music 3
MiniMax AI · Music Generation · Week 33, 2026
- VocalRender
pymaster17 · Music Generation · Week 32, 2026
- SymphonyGen
Central Conservatory of Music, XJTU, USTC & Shenzhen University · Music Generation · Week 32, 2026
- AudioLM
Google Research (Borsos, Marinier, Vincent, Kharitonov, Pietquin, Sharifi, Roblek, Teboul, Grangier, Tagliasacchi, Zeghidour) · Music Generation · Week 30, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.