zzz66 commited on
Commit
96465d6
·
1 Parent(s): 63a1b7e

Use precompiled flash_attn wheel

Browse files
Files changed (1) hide show
  1. requirements.txt +3 -2
requirements.txt CHANGED
@@ -26,5 +26,6 @@ onnxruntime-gpu==1.22.0
26
  insightface==0.7.3
27
  transformers==4.52.0
28
  huggingface_hub
29
- ninja
30
- flash_attn==2.8.1
 
 
26
  insightface==0.7.3
27
  transformers==4.52.0
28
  huggingface_hub
29
+ ninja
30
+ # 使用预编译的 flash_attn wheel (torch 2.6.0 + CUDA 12.4 + Python 3.10)
31
+ https://github.com/Dao-AILab/flash-attention/releases/download/v2.8.1/flash_attn-2.8.1+cu124torch2.6cxx11abiFALSE-cp310-cp310-linux_x86_64.whl