# riffusion/riffusion-hobby

Stable diffusion for real-time music generation

Repository: https://github.com/riffusion/riffusion-hobby
Canonical: https://ross.abutalabs.com/products/riffusion-hobby
Language: Python
License: MIT
License Family: permissive
Topics: diffusion, ai, audio, music, diffusers, stable-diffusion
Last push: 2024-07-22T18:09:39+00:00

## Health v2 (maintenance only)
Score: 22/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 98
- inputs: {"age_days": 1377, "days_push": 772, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3900, forks 479 (observed 2026-08-28T04:08:28.586992+00:00)

## What it is
Riffusion is a Python library for real-time music and audio generation using stable diffusion, converting between spectrogram images and audio clips with prompt interpolation. It includes a CLI, a Streamlit interactive app, and a Flask inference server.

## Use cases
- generate music from text prompts with stable diffusion
- convert spectrogram images to audio clips
- run real-time audio generation on a GPU
- serve a diffusion model inference API
- experiment with prompt interpolation for audio

## When to choose
- you want to generate music or audio from text prompts locally
- you need spectrogram-to-audio conversion utilities in Python
- you have a CUDA GPU capable of running stable diffusion
- you want to build on or study a diffusion-based audio pipeline

## When to avoid
- you need an actively maintained project - it is no longer maintained
- you only have a CPU and need real-time generation
- you want a hosted or production-ready music service
- you need non-WAV audio without installing ffmpeg

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, audio-processing, image-processing, cli, http-server
- domain: artificial-intelligence, machine-learning, deep-learning, media
- platform: python, cross-platform, cli
- tags: stable-diffusion, music-generation, spectrogram, diffusion-models, text-to-audio, generative-ai, streamlit, flask, audio, gpu

## Member repositories
- riffusion/riffusion-hobby (main) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:28.586992+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:25:11.712468+00:00, confidence not recorded.
  - readme: https://github.com/riffusion/riffusion-hobby (fetched 2026-08-28T04:08:28.586992+00:00, sha f8de850f6b37)
- Data as of 2026-08-30T08:39:29.467469+00:00.