ggml-cuda: fix divergent barrier in f16 flash attention #27870
ggml-cuda: fix divergent barrier in f16 flash attention
672f59d8
ggml-cuda: avoid duplicate metadata pointer setup
c4b9b7e1
ggerganov
merged
b74f590e
into master 76 days ago
Labels
ggml
merge ready
CUDA
Login to write a write a comment.
Login via GitHub