SongGeneration 2 (LeVo 2)
Released 2026-03-01 (per vLLM-Omni model request citing the official repo). Reported lyric accuracy PER 8.55% vs Suno v5 12.4% and Mureka v8 9.96% (secondary source gaga.art, not verified). As of 2026-09-29 the official GitHub repo github.com/tencent-ailab/SongGeneration returns 404 and the tencent/SongGeneration HF repo returns 401 (apparently removed/made private; community forks and reuploads exist, e.g. Pinokio notes); lglg666/SongGeneration-v2-large (created 2026-02-15, license 'unknown') is still public. Treat availability and license as unverified. Demo: https://levo-demo.github.io/levo_v2_demo/
- Input
- text, audio
- Output
- audio
- License
- custom Tencent terms (GitHub showed NOASSERTION; not verified)
How to call it
| Provider | Model id | Endpoint / URL | Docs |
|---|---|---|---|
| Hugging Face (v2-large checkpoint, uploader account) | — | huggingface.co/lglg666/SongGeneration-v2-large | — |
| Hugging Face (official org repo; returned 401 on 2026-09-29) | — | huggingface.co/tencent/SongGeneration | — |
Notable capabilities (2)
- Hybrid LLM-diffusion full songs up to 4:30: 4B-parameter model generating complete songs up to 4 min 30 s with vocals + accompaniment, instrumental-only, a cappella or dual-track (separated) output; multilingual lyrics (Chinese, English, Spanish, Japanese and more). source
- Hierarchical semantic planning + track-specific refinement (found after launch): LeVo 2 paper: semantic planning precedes per-track refinement to keep vocal-instrument coordination while improving acoustics; progressive post-training with automatic quality tiers. source
Tencent's open song model family (v1 LeVo, arXiv 2506.07520; v2 LeVo 2, arXiv 2606.30642). Availability is uncertain after the official repos went offline.