Duplicated from Ketengan-Diffusion-Lab/Dolphin-Inference
e6de351 f3c7bef fca9b9b 0ebbca3 fd57ab6 fc86b40 fd57ab6 fca9b9b
1
2
3
4
5
6
7
8
9
wheel ninja git+https://github.com/Dao-AILab/flash-attention.git transformers gradio spaces decord einops sentencepiece