Add flash-attention 2 for windows #4235
kingbri1
force pushed
from
82fb209b
to
722fac52
2 years ago
Requirements: Add cuda 12.1 and update one click
06ad1450
kingbri1
force pushed
from
722fac52
to
06ad1450
2 years ago
Bump to CUDA 12.1 & Python 3.11
0b714d71
Merge branch 'main' into bdashore3-flash-attention-windows
cef34a83
Change choise to "Would you like CUDA 11.8?"
9bc14062
Minor fixes
0f25ee56
Merge branch 'main' into bdashore3-flash-attention-windows
b4f56533
Don't use python 3.11
8e7d7655
Minor change
83140c3f
Update README
e7c662ee
CUDA_118 environment variable
991a9ff9
Fix typo
63f4c344
More foolproof
eb4cc7d6
cu118 -> cu121
81f9112e
cu122 for flash-attn
2d0ea337
Add new wheels
5a61197e
Use python 3.11
df9966d5
More robust flash-attention import
cd33585f
Lint
ec9ddeeb
Update dockerfile
8465893f
oobabooga
merged
3345da2e
into main 2 years ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub