Diffusion-Pipe In ComfyUI Custom Nodes

Attention! This is the Linux(wsl2) version

If you use this custom node on Windows, you will not be able to train

注意！此处为Linux(wsl2)版本

如果你在windows上使用了这个插件，你将无法训练

点击查看 中文文档

Project Overview

Diffusion-Pipe In ComfyUI custom nodes is a powerful extension plugin that provides complete Diffusion model training and fine-tuning functionality for ComfyUI. This project allows users to configure and launch training for various advanced AI models through ComfyUI's graphical interface, supporting both LoRA and full fine-tuning, covering the most popular image generation and video generation models.

Video Demo: https://www.bilibili.com/video/BV1CRk9BYErw/?vd_source=7fd137e57a445e84bd9ffea9b632c98d

Please read the notice in the workflow before training

Quick Start

update

• 20260117: flux2 support, and follow the super huge update of the original project，deepspeed updated，comfyui submodules updated

• 20260103: add qwen2512 support, same as qwen_image

• 20260102: qwen2511 support, supports both Diffusers and ComfyUI format models

• 20251231, Zimage FP8 Comfy format training, supported by hunyuan-video_15 model.

• 20251130: Z-Image support, supports both Diffusers and ComfyUI format models

You need to download the latest diffusers development version to support training, e.g.:

 pip install git+https://github.com/huggingface/diffusers

Z-Image-Turbo

merge_adapters = ['/data2/imagegen_models/comfyui-models/zimage_turbo_training_adapter_v1.safetensors']

Model files support using the ComfyUI version.

Also supports diffusers

If training Z-Image-Turbo, make sure to merge the adapter.

Credit to Ostris and AI Toolkit for making this adapter.

Z-Image LoRAs are saved in ComfyUI format. This is different from Diffusers format.

• 20251026:support eval

• 20251030:Supports training Aura models

• 20251103:support MultiImage Edit (qwen2509)

• 20251105:support mask trainning，Fix off-by-one error in plots when using examples as x-axis,Allow using captions.json without tar files,add reset_optimizer flag,--reset_optimizer_params flag（Reset optimizer parameters, which allows resetting the optimizer during resuming training）,Fix datasets issue,Cast to float16 in dataset caching to cut size on disk in half

Installation Guide

Installation

Make sure you have ComfyUI on Linux or WSL2 system, refer to https://docs.comfy.org/installation/manual_install

ps: ComfyUI on WSL2 works so well that I even want to delete my ComfyUI on Windows

conda create -n comfyui_DP python=3.12

conda activate comfyui_DP

cd ~/comfy/ComfyUI/custom_nodes/

git clone --recurse-submodules https://github.com/TianDongL/Diffusion_pipe_in_ComfyUI.git

• If you haven't installed submodules, follow these steps

• If you don't do this step, training will not work

git submodule init
git submodule update

Install Dependencies

conda activate comfyui_DP

Here are the necessary dependencies for deepspeed, first install PyTorch. It is not listed in the requirements document because some GPUs sometimes require different versions of PyTorch or CUDA, and you may have to find a combination that suits your hardware.

pip install torch==2.7.1 torchvision==0.22.1 torchaudio==2.7.1 --index-url https://download.pytorch.org/whl/cu128

cd ~/comfy/ComfyUI/custom_nodes/Diffusion_pipe_in_ComfyUI

pip install -r requirements.txt

🚀 One-Click Import Workflow

To get you started quickly, we provide pre-configured ComfyUI workflow files:

📋 Click to Import Complete Workflow

Drag this file into the ComfyUI interface to import the complete training workflow, including all necessary node configurations.

Please read the prompts in the workflow carefully, this can help you build your dataset

📷 Workflow Interface Preview

Core Features

• 🎯 Visual Training Configuration: Graphically configure training parameters through ComfyUI nodes
• 🚀 Multi-Model Support: Support for 20+ latest Diffusion models
• 💾 Flexible Training Methods: Support for LoRA training and full fine-tuning
• ⚡ High-Performance Training: DeepSpeed-based distributed training support
• 📊 Real-time Monitoring: Integrated TensorBoard training process monitoring
• 🔧 WSL2 Optimization: Specially optimized Windows WSL2 environment support
• 🎥 Video Training: Support for video generation model training
• 🖼️ Image Editing: Support for image editing model training

System Requirements

Hardware Requirements

• • I don't know, you can try :-P

Software Requirements

• Operating System: Linux / Windows 10/11 + WSL2
• ComfyUI: Latest version

Supported Models

This plugin supports over 20 latest Diffusion models, including:

License

This project is open source under the Apache License 2.0.

Contributing Guide

Issues and Pull Requests are welcome!

1. Fork the project
2. Create a feature branch
3. Submit changes
4. Create a Pull Request

Acknowledgments

Thanks to the following projects and teams:

• ComfyUI team
• @tdrussell, the original author of Diffusion_Pipe
• Hugging Face Diffusers
• DeepSpeed team
• Original authors of various models