diff --git a/TeaCache4Lumina2/README.md b/TeaCache4Lumina2/README.md new file mode 100644 index 0000000..9cd0e8e --- /dev/null +++ b/TeaCache4Lumina2/README.md @@ -0,0 +1,49 @@ + +# TeaCache4LuminaT2X + +[TeaCache](https://github.com/LiewFeng/TeaCache) can speedup [Lumina-Image-2.0](https://github.com/Alpha-VLLM/Lumina-Image-2.0) 2x without much visual quality degradation, in a training-free manner. The following image shows the results generated by TeaCache-Lumina-Image-2.0 with various rel_l1_thresh values: 0 (original), 0.1 (1.05x speedup), 0.2 (1.15x speedup), 0.3 (1.25x speedup). + +

+ + + + +

+ +## 📈 Inference Latency Comparisons on a 4070 laptop(size 1024 x 1536) + + +| Lumina-Image-2.0 | TeaCache (0.1) | TeaCache (0.2) | TeaCache (0.3) | +|:---------------------------:|:-----------------------------:|:--------------------:|:---------------------:| +| ~97.74s | ~93.19s | ~84.72s | ~78.43s | + +## Installation + +```shell +pip install --upgrade diffusers[torch] transformers protobuf tokenizers sentencepiece +pip install flash-attn --no-build-isolation +``` + +## Usage + +You can modify the thresh in line 113 to obtain your desired trade-off between latency and visul quality. For single-gpu inference, you can use the following command: + +```bash +python teacache_lumina2.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 [Lumina-Image-2.0](https://github.com/Alpha-VLLM/Lumina-Image-2.0) and [Diffusers](https://github.com/huggingface/diffusers).