Machine Learning Library
33.5k 2026-04-30
huggingface/diffusers
A modular PyTorch library for state-of-the-art diffusion models, enabling easy inference and training for image, video, and audio generation.
Core Features
State-of-the-art diffusion pipelines for inference with minimal code.
Interchangeable noise schedulers for diverse diffusion speeds and output quality.
Pretrained models as flexible building blocks for custom diffusion systems.
Designed with a strong focus on usability, simplicity, and customizability.
Quick Start
pip install --upgrade diffusers[torch]Detailed Introduction
Hugging Face Diffusers is a comprehensive PyTorch library designed to democratize access to advanced diffusion models. It provides a modular toolkit for both rapid inference and custom model training, supporting the generation of high-quality images, videos, and audio. With its emphasis on usability and customizability, Diffusers empowers developers and researchers to easily integrate and experiment with cutting-edge generative AI, offering a flexible foundation for building novel applications and exploring the frontiers of AI-driven content creation.