Diffusers is Hugging Face's open-source Python library for running and training diffusion models that generate images, video and audio.
In plain words
Diffusers wraps each model in a pipeline you can run with a few lines of code. The parts are interchangeable: you can swap the noise scheduler to trade speed for quality, add LoRA adapters, or use memory-saving tricks such as offloading and quantization on smaller GPUs.
A quick example
Imagine a hobbyist on a laptop GPU who wants to try an open image model. They load its pipeline from the Hugging Face Hub, turn on offloading so it fits in memory, and generate a picture from a text prompt.
Where you'll meet it
- Running open image and video models from Python.
- Fine-tuning diffusion models, for example with LoRA.
Related
Sources
- DocsDiffusers, Hugging Face · read 6 Oct 2026
- DocsDiffusers README, Hugging Face (GitHub) · read 6 Oct 2026
- Docshuggingface/diffusers, Hugging Face (GitHub) · read 6 Oct 2026
Checked by WhatIsAI against the sources above · 6 Oct 2026. Spotted an error? Report it.