https://github.com/smthemex/ComfyUI_VisualCloze
This node is base on VisualCloze method, A Universal Image Generation Framework via Visual In-Context Learning
https://github.com/smthemex/ComfyUI_VisualCloze
Last synced: 7 days ago
JSON representation
This node is base on VisualCloze method, A Universal Image Generation Framework via Visual In-Context Learning
- Host: GitHub
- URL: https://github.com/smthemex/ComfyUI_VisualCloze
- Owner: smthemex
- License: mit
- Created: 2025-04-23T07:27:18.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2025-05-21T08:56:45.000Z (29 days ago)
- Last Synced: 2025-05-21T09:52:35.381Z (29 days ago)
- Language: Python
- Size: 89.6 MB
- Stars: 10
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-comfyui - **ComfyUI_VisualCloze** - Context Learning,you can use it in ComfyUI (All Workflows Sorted by GitHub Stars)
README
[VisualCloze](https://github.com/lzyhha/VisualCloze): A Universal Image Generation Framework via Visual In-Context Learning
# Tips
* 量化fp8的模式始终跑不出效果，官方的diffuser版本暂时没空捣鼓，先放代码出来，免得说占坑。方法需要的显存较大，内存也要64+,24显存可以试试关闭cpu offload
1.Installation
-----
In the ./ComfyUI /custom_node directory, run the following:
```
git clone https://github.com/smthemex/ComfyUI_VisualCloze.git
```
2.requirements
----
```
pip install -r requirements.txt
```
3 models
----
[lora 384](https://huggingface.co/VisualCloze/VisualClozePipeline-LoRA-384/tree/main) or [lora 512](https://huggingface.co/VisualCloze/VisualClozePipeline-LoRA-512/tree/main)
```
├── ComfyUI/models/loras/
| ├── visualcloze-lora-512.safetensors # or 384
├── ComfyUI/models/diffusion_models/
| ├── flux1-fill-dev.safetensors # or flux1-fill-dev-fp8.safetensors
```
# Example
# Citation
```
@article{li2025visualcloze,
title={VisualCloze : A Universal Image Generation Framework via Visual In-Context Learning},
author={Li, Zhong-Yu and Du, Ruoyi and Yan, Juncheng and Zhuo, Le and Li, Zhen and Gao, Peng and Ma, Zhanyu and Cheng, Ming-Ming},
journal={arXiv preprint arxiv:},
year={2025}
}
```