https://github.com/zichongc/ComfyUI-Attention-Distillation
Official Implementation of Attention Distillation for ComfyUI
https://github.com/zichongc/ComfyUI-Attention-Distillation
Last synced: about 1 month ago
JSON representation
Official Implementation of Attention Distillation for ComfyUI
- Host: GitHub
- URL: https://github.com/zichongc/ComfyUI-Attention-Distillation
- Owner: zichongc
- License: mit
- Created: 2025-02-27T09:19:56.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2025-02-27T09:27:47.000Z (about 2 months ago)
- Last Synced: 2025-02-27T13:06:57.729Z (about 2 months ago)
- Language: Python
- Size: 688 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-comfyui - **ComfyUI-Attention-Distillation** - native [a/AttentionDistillation](https://github.com/xugao97/AttentionDistillation) for ComfyUI. (All Workflows Sorted by GitHub Stars)
README
## ComfyUI-Attention-Distillation
Non-native [AttentionDistillation](https://github.com/xugao97/AttentionDistillation) for ComfyUI.
Official ComfyUI demo for the paper [AttentionDistillation](), implemented as an extension of ComfyUI. Note that this extension incorporates AttentionDistillation using `diffusers`.
The official code for AttentionDistillation can be found [here]().
### 🔥🔥 News
* **2025/02/27**: We release the ComfyUI implementation of Attention Distillation and two workflows for style transfer and style-specific text-to-image generation using Stable Diffusion 1.5.
* **2025/02/27**: The official code for AttentionDistillation has been released [here](https://github.com/xugao97/AttentionDistillation).
### 🛒 Installation
Download or `git clone` this repository into the `ComfyUI/custom_nodes/` directory, or use the Manager for a streamlined setup.
##### Install manually
1. `cd custom_nodes`
2. `git clone ...`
3. `cd custom_nodes/ComfyUI-AttentionDistillation`
4. `pip install -r requirements.txt`
5. restart ComfyUI
### 📒 How to Use
##### Download T2I diffusion models
This implementation utilizes checkpoints for `diffusers`. Download the required models and place them in the `ComfyUI/models/diffusers` directory:
|Model|Model Name and Link|
|:---:|:---:|
| Stable Diffusion (v1.5, v2.1) | [stable-diffusion-v1-5/stable-diffusion-v1-5](https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5)
[stabilityai/stable-diffusion-2-1](https://huggingface.co/stabilityai/stable-diffusion-2-1) |
| SDXL | [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) |
*Note: Currently, only Stable Diffusion v1.5 is required.*
##### Load the workflow
Workflows for various tasks are available in `ComfyUI/custom_nodes/Comfy-Attention-Distillation/workflows`. Simply load them to get started. Additionally, we've included usage examples in the [Examples](#examples) section for your reference.
### 🔍 Examples
#### Style-specific text-to-image generation
`style_t2i_generation_sd15.json`
#### Style Transfer
`style_transfer_sd15.json`
### 📃 TODOs
- [ ] Workflow for style-specific text-to-image generation using SDXL.
- [ ] Workflow for texture synthesis.