Releases: Dao-AILab/flash-attention
Releases · Dao-AILab/flash-attention
v2.3.2
Bump to v2.3.2
v2.3.1.post1
[CI] Use official Pytorch 2.1, add CUDA 11.8 for Pytorch 2.1
v2.3.1
Bump to v2.3.1
v2.3.0
Bump to v2.3.0
v2.2.5
Bump to v2.2.5
v2.2.4.post1
Re-enable compilation for Hopper
v2.2.4
Bump to v2.2.4
v2.2.3.post2
Don't compile for Pytorch 2.1 on CUDA 12.1 due to nvcc segfaults
v2.2.3.post1
Set block size to 64 x 64 for kvcache to avoid nvcc segfaults
v2.2.3
Bump to v2.2.3