DiffRhythm

Spec sheet

What it does

DiffRhythm is an open-source diffusion model for generating full-length songs.Read 12 Oct 2026 · github.com

Purpose

DiffRhythm is an open-source diffusion music model designed to generate full-length songs.Read 12 Oct 2026 · github.com

Intended users

The project describes itself as a foundation for research and development and says it supports artistic creation, education, and entertainment.Read 12 Oct 2026 · github.com

Deployment

The README documents local inference and Docker deployment, including setup instructions for Linux, macOS, and Windows.Read 12 Oct 2026 · github.com

Support

The repository invites messages to the research team at [email protected].Read 12 Oct 2026 · github.com

License

The code and DiT weights are released under Apache License 2.0, which allows use, modification, and distribution subject to its notices and disclaimer.Read 12 Oct 2026 · github.com

Audio export formats

wavRead 21 Sep 2026 · github.com

Audio upload

YesRead 21 Sep 2026 · github.com

Commercial use

YesRead 21 Sep 2026 · github.com

Demo

The project links to a Hugging Face Space demo and a separate project demo page.Read 12 Oct 2026 · github.com

Editing

The project says version 1.2 enables song editing and continuation and improves its handling of music structure and style.Read 12 Oct 2026 · github.com

Hardware requirement

DiffRhythm-base requires at least 8 GB of VRAM; the repository says chunked decoding can help meet this requirement.Read 12 Oct 2026 · github.com

Input modes

The project describes text-based style prompts, instrumental generation, and generation using a reference WAV file.Read 12 Oct 2026 · github.com

Inputs and modes

It supports text-based style prompts, instrumental generation, and inference using a reference WAV or a text prompt reference.Read 12 Oct 2026 · github.com

Installation

The repository provides local inference instructions for Linux, macOS, and Windows, and includes Docker installation instructions.Read 12 Oct 2026 · github.com

Pricing

The project says its makers do not make a profit from the model; the page lists no paid plans or prices.Read 12 Oct 2026 · github.com

Responsible use

The makers advise users to verify musical originality, disclose AI involvement, and obtain permissions when adapting protected styles.Read 12 Oct 2026 · github.com

Song length

The listed base models generate 1 minute 35 second songs, and the full models generate 4 minute 45 second songs.Read 12 Oct 2026 · github.com

Training

The repository says training is coming soon.Read 12 Oct 2026 · github.com

Training code

The README lists training as coming soon.Read 12 Oct 2026 · github.com

Vocal generation

YesRead 21 Sep 2026 · github.com

Where Specifiction places it

• AI music generators#91 of 1237.5