# DiffRhythm

> DiffRhythm from ASLP-lab generates songs with vocals from LRC-timed lyrics plus a text style prompt or a reference clip, with an instrumental mode; v1.2 (May 2025) adds song editing and continuation. Base models make 1m35s, full models 4m45s. It needs about 8 GB of VRAM with chunked decoding and espeak-ng installed. Code and DiT weights are Apache-2.0, but the VAE uses the Stability AI Community licence. There is no Gradio UI; you run shell scripts. Last code push was November 2025, and DiffRhythm 2 is announced elsewhere.

- Repository: https://github.com/ASLP-lab/DiffRhythm
- Official docs: https://github.com/ASLP-lab/DiffRhythm#readme
- Checked: 2026-10-15

## What it replaces
- Suno: Full songs with vocals from a prompt or lyrics, Instrumental tracks (Instrumental mode), Extending or regenerating part of a song (Extend (partial). Songs, instrumentals and continuation from scripts; lyrics must be timed LRC; no covers or stems.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Yes, inference scripts or Docker on your own machine | https://github.com/ASLP-lab/DiffRhythm#inference | 2026-10-15 |
| Code licence | Confirmed | Apache-2.0 | https://github.com/ASLP-lab/DiffRhythm#license--disclaimer | 2026-10-15 |
| Model weights licence | Partial | DiT weights Apache-2.0; the DiffRhythm VAE is under the Stability AI Community licence | https://huggingface.co/ASLP-lab/DiffRhythm-vae | 2026-10-15 |
| Commercial use | Partial | Free for commercial use under USD 1M annual revenue because of the VAE's Stability AI Community licence; larger firms need a Stability licence | https://stability.ai/license | 2026-10-15 |
| Paid services needed? | Confirmed | None required | https://github.com/ASLP-lab/DiffRhythm#inference | 2026-10-15 |
| Hardware | Confirmed | At least 8 GB VRAM with the --chunked option; more without it | https://github.com/ASLP-lab/DiffRhythm#inference | 2026-10-15 |
| Operating systems | Confirmed | Linux, Windows (batch scripts, espeak-ng setup) and macOS | https://github.com/ASLP-lab/DiffRhythm#inference | 2026-10-15 |
| Setup effort | Confirmed | Medium; conda or venv plus espeak-ng, or Docker compose | https://github.com/ASLP-lab/DiffRhythm#docker-installation | 2026-10-15 |

GitHub: 2347 stars, 278 forks, last push 2025-11-27, archived: no.
---
Source: RepoQuarry (https://repoquarry.dev/projects/diffrhythm). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.