As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/songgeneration-2/

# SongGeneration 2 (LeVo 2)

Tencent AI Lab, SongGeneration / LeVo family. Status: current. Type: music generation. Released: 2026-03-01.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Hugging Face (v2-large checkpoint, uploader account) |  |  | https://huggingface.co/lglg666/SongGeneration-v2-large |
| Hugging Face (official org repo; returned 401 on 2026-09-29) |  |  | https://huggingface.co/tencent/SongGeneration |

## Pricing

- Price not published.
- Last checked: not recorded

## Specs

- Input: text, audio
- Output: audio
- Context window: not published
- Max output: not published
- Open weights: yes
- Licence: custom Tencent terms (GitHub showed NOASSERTION; not verified)

## Where it stands

- Newer music generation models in the registry since its release: 7. The 5 newest:
  - Eleven Music v2.5 (2026-09-11) https://postcutoff.com/m/elevenlabs-music-v2-5/
  - Suno v6 (v6, v6-wild, v6-mini) (2026-09-09) https://postcutoff.com/m/suno-v6/
  - YuE2-3B (2026-09-09) https://postcutoff.com/m/yue2/
  - Lyria 3.5 (2026-07-29) https://postcutoff.com/m/lyria-3-5/
  - MiniMax Music 3.0 (2026-07-16) https://postcutoff.com/m/minimax-music-3/
- Weights are public. Licence: custom Tencent terms. See the licence for commercial terms.

## What stands out

- 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: https://github.com/vllm-project/vllm-omni/issues/3390
- 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: https://arxiv.org/abs/2606.30642

## Notes

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/

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.

Model page: https://postcutoff.com/m/songgeneration-2/