# modelscope/DiffSynth-Studio

Enjoy the magic of Diffusion models!

Repository: https://github.com/modelscope/DiffSynth-Studio
Canonical: https://ross.abutalabs.com/products/diffsynth-studio
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-26T06:16:01+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 57, longevity 71
- inputs: {"age_days": 1000, "days_push": 7, "days_rel": 289, "gap_med": 25, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 13003, forks 1283 (observed 2026-08-28T04:11:00.366391+00:00)

## What it is
DiffSynth-Studio is an open-source diffusion model engine from the ModelScope community that integrates mainstream image, video, and audio generation models. It provides VRAM management, parameter quantization, and flexible training support for base models, LoRAs, and adapters.

## Use cases
- generate images from text prompts with diffusion models
- generate videos from text or images
- train LoRA adapters for diffusion models on consumer GPUs
- run large diffusion models on low-VRAM graphics cards
- quantize diffusion models to NF4 or INT8
- generate audio with diffusion models
- fine-tune image generation models

## When to avoid
- you need a simple end-user GUI app rather than a Python framework
- you only need non-diffusion generative models like plain LLMs
- you require production serving infrastructure with APIs out of the box

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, deep-learning, image-processing, video-processing, audio-processing, llm-training, gpu-computing
- domain: deep-learning, machine-learning, image-processing, artificial-intelligence
- platform: python, windows
- tags: diffusion-models, image-generation, video-generation, lora, model-quantization, vram-management, text-to-image, text-to-video, modelscope, video, audio, gpu, linux, macos

## Member repositories
- modelscope/DiffSynth-Studio (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:00.366391+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 2026-08-29T17:13:36.527749+00:00, confidence not recorded.
  - readme: https://github.com/modelscope/DiffSynth-Studio (fetched 2026-08-28T04:11:00.366391+00:00, sha 7c2e05f80f61)
- Data as of 2026-08-30T08:39:29.467469+00:00.