# ComfyUI-Diffusers-X-Adapter

**Repository Path**: analyzesystem/ComfyUI-Diffusers-X-Adapter

## Basic Information

- **Project Name**: ComfyUI-Diffusers-X-Adapter
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2024-03-19
- **Last Updated**: 2024-04-16

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# ComfyUI wrapper node for X-Adapter diffusers implementation.

This is meant for testing only, with the ability to use same models and python env as ComfyUI, it is NOT a proper ComfyUI implementation!
### I won't be bothering with backwards compability with this node, in many updates you will have to remake any existing nodes (or set widget values again)

# Known limitations:
  - As this is only a wrapper, it's not compatible with anything else in ComfyUI, besides input preprocessing and being able to load and convert most models for the Diffusers pipeline
  - Ohe ratio between 1.5 and SDXL resolution also has to be exactly 1:2
  - Some ControlNets/LoRAs won't load, and results with some combos seem broken

![Screenshot from 2024-02-22 14-15-35](https://github.com/kijai/ComfyUI-Diffusers-X-Adapter/assets/40791699/8a101ed1-e00c-41d3-98bd-b5cf38dc6123)
![image](https://github.com/kijai/ComfyUI-Diffusers-X-Adapter/assets/40791699/bf9dc281-8f9f-43d3-99c5-3ba255ea69bf)
![image](https://github.com/kijai/ComfyUI-Diffusers-X-Adapter/assets/40791699/530bb971-a8e5-4688-94d4-d7949dd56e1d)

# X-Adapter

This repository is the official implementation of [X-Adapter](https://arxiv.org/abs/2312.02238).

**[X-Adapter: Adding Universal Compatibility of Plugins for Upgraded Diffusion Model](https://arxiv.org/abs/2312.02238)**
<br/>
[Lingmin Ran](),
[Xiaodong Cun](https://vinthony.github.io/academic/),
[Jia-Wei Liu](https://jia-wei-liu.github.io/), 
[Rui Zhao](https://ruizhaocv.github.io/), 
[Song Zijie](), 
[Xintao Wang](https://xinntao.github.io/),
[Jussi Keppo](https://www.jussikeppo.com/), 
[Mike Zheng Shou](https://sites.google.com/view/showlab)
<br/>

[![Project Website](https://img.shields.io/badge/Project-Website-orange)](https://showlab.github.io/X-Adapter/)
[![arXiv](https://img.shields.io/badge/arXiv-2312.02238-b31b1b.svg)](https://arxiv.org/abs/2312.02238)

![Overview_v7](https://github.com/showlab/X-Adapter/assets/152716091/eb41c508-826c-404f-8223-09765765823b)

<em> X-Adapter enable plugins pretrained on old version (e.g. SD1.5) directly work with the upgraded Model (e.g., SDXL) without further retraining.</em>

[//]: # (<p align="center">)

[//]: # (<img src="https://tuneavideo.github.io/assets/teaser.gif" width="1080px"/>  )

[//]: # (<br>)

[//]: # (<em>Given a video-text pair as input, our method, Tune-A-Video, fine-tunes a pre-trained text-to-image diffusion model for text-to-video generation.</em>)

[//]: # (</p>)


## Cite
If you find X-Adapter useful for your research and applications, please cite us using this BibTeX:

```bibtex
@article{ran2023xadapter,
  title={X-Adapter: Adding Universal Compatibility of Plugins for Upgraded Diffusion Model},
  author={Lingmin Ran and Xiaodong Cun and Jia-Wei Liu and Rui Zhao and Song Zijie and Xintao Wang and Jussi Keppo and Mike Zheng Shou},
  journal={arXiv preprint arXiv:2312.02238},
  year={2023}
}
```