Diffusers

Get started

🧨 Diffusers Quicktour Effective and efficient diffusion Installation

Tutorials

Overview Understanding pipelines, models and schedulers AutoPipeline Train a diffusion model

Load pipelines and adapters

Load pipelines Load community pipelines and components Load schedulers and models Model files and layouts Push files to the Hub

Adapters

Generative tasks

Unconditional image generation Text-to-image Image-to-image Inpainting Video generation Depth-to-image

Inference techniques

Overview Create a server Distributed inference Scheduler features Pipeline callbacks Reproducible pipelines Controlling image quality Prompt techniques

Advanced inference

Outpainting

Hybrid Inference

Overview VAE Decode VAE Encode API Reference

Specific pipeline examples

ConsisID Stable Diffusion XL SDXL Turbo Kandinsky OmniGen PAG Latent Consistency Model Shap-E DiffEdit Trajectory Consistency Distillation-LoRA Stable Video Diffusion Marigold Computer Vision

Training

Overview Create a dataset for training Adapt a model to a new task

Models

Methods

Quantization Methods

Getting Started bitsandbytes gguf torchao quanto

Accelerate inference and reduce memory

Accelerate inference Caching Reduce memory usage Compile and offloading quantized models Pruna xFormers Token merging DeepCache TGATE xDiT ParaAttention

Optimized model formats

JAX/Flax ONNX OpenVINO Core ML

Optimized hardware

Metal Performance Shaders (MPS) Intel Gaudi AWS Neuron

Conceptual Guides

Philosophy Controlled generation How to contribute? Diffusers' Ethical Guidelines Evaluating Diffusion Models

Community Projects

Projects built with Diffusers

API

Main Classes

Loaders

Models

Pipelines

Schedulers

Internal classes

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

500

Not Found

←AutoPipeline Load pipelines→

Train a diffusion model Training configuration Load the dataset Create a UNet2DModel Create a scheduler Train the model Next steps