mirror of
https://git.datalinker.icu/ali-vilab/TeaCache
synced 2025-12-08 20:34:24 +08:00
43 lines
1.9 KiB
Markdown
43 lines
1.9 KiB
Markdown
<!-- ## **TeaCache4FLUX** -->
|
|
# TeaCache4FLUX
|
|
|
|
[TeaCache](https://github.com/LiewFeng/TeaCache) can speedup [FLUX](https://github.com/black-forest-labs/flux) 2x without much visual quality degradation, in a training-free manner. The following image shows the results generated by TeaCache-FLUX with various `rel_l1_thresh` values: 0 (original), 0.25 (1.5x speedup), 0.4 (1.8x speedup), 0.6 (2.0x speedup), and 0.8 (2.25x speedup).
|
|
|
|
|
|
|
|
## 📈 Inference Latency Comparisons on a Single A800
|
|
|
|
|
|
| FLUX.1 [dev] | TeaCache (0.25) | TeaCache (0.4) | TeaCache (0.6) | TeaCache (0.8) |
|
|
|:-----------------------:|:----------------------------:|:--------------------:|:---------------------:|:---------------------:|
|
|
| ~18 s | ~12 s | ~10 s | ~9 s | ~8 s |
|
|
|
|
## Installation
|
|
|
|
```shell
|
|
pip install --upgrade diffusers[torch] transformers protobuf tokenizers sentencepiece
|
|
```
|
|
|
|
## Usage
|
|
|
|
You can modify the `rel_l1_thresh` in line 320 to obtain your desired trade-off between latency and visul quality. For single-gpu inference, you can use the following command:
|
|
|
|
```bash
|
|
python teacache_flux.py
|
|
```
|
|
|
|
## Citation
|
|
If you find TeaCache is useful in your research or applications, please consider giving us a star 🌟 and citing it by the following BibTeX entry.
|
|
|
|
```
|
|
@article{liu2024timestep,
|
|
title={Timestep Embedding Tells: It's Time to Cache for Video Diffusion Model},
|
|
author={Liu, Feng and Zhang, Shiwei and Wang, Xiaofeng and Wei, Yujie and Qiu, Haonan and Zhao, Yuzhong and Zhang, Yingya and Ye, Qixiang and Wan, Fang},
|
|
journal={arXiv preprint arXiv:2411.19108},
|
|
year={2024}
|
|
}
|
|
```
|
|
|
|
## Acknowledgements
|
|
|
|
We would like to thank the contributors to the [FLUX](https://github.com/black-forest-labs/flux) and [Diffusers](https://github.com/huggingface/diffusers). |