diffsynth

Documentation Introduction

  • DiffSynth-Studio Documentation

Getting Started

  • Installing Dependencies
  • Model Inference
  • Inference Acceleration
  • VRAM Management
  • Model Training
  • Environment Variables
  • GPU/NPU Support
  • Inference WebUI

Model Details

  • FLUX
  • Wan
  • Qwen-Image
  • FLUX.2
  • Z-Image
  • Anima
  • LTX-2
  • ERNIE-Image
  • JoyAI-Image
  • ACE-Step
  • HiDream-O1-Image
  • Stable Diffusion
  • Stable Diffusion XL
  • Image Quality Evaluation Metrics
  • Ideogram 4

Training Framework

  • Basic Principles of Diffusion Models
  • Standard Supervised Training
  • Enabling FP8 Precision in Training
  • End-to-End Distillation Accelerated Training
  • Two-Stage Split Training
  • Differential LoRA Training
  • Enabling DeepSpeed
  • Offload Training

Model Integration

  • Integrating Model Architecture
  • Building a Pipeline
  • Fine-Grained VRAM Management Scheme
  • Integrating Model Training

API Reference

  • diffsynth.core.attention: Attention Mechanism Implementation
  • diffsynth.core.data: Data Processing Operators and Universal Dataset
  • diffsynth.core.gradient: Gradient Checkpointing and Offload
  • diffsynth.core.loader: Model Download and Loading
  • diffsynth.core.vram: VRAM Management

Diffusion Templates

  • Diffusion Templates
  • Diffusion Templates Architecture Details
  • Template Model Inference
  • Template Model Training

Research Guide

  • Training Models from Scratch
  • Inference Optimization Techniques

FAQ

  • Frequently Asked Questions
diffsynth
  • Search

© Copyright 2022-2025, Alibaba ModelScope.

Built with Sphinx using a theme provided by Read the Docs.