david/aphrodite-engine

mirrorاز https://github.com/PygmalionAI/aphrodite-engine

نویسنده	SHA1 پیام	تاریخ
AlpinDale	f8dfac6372 chore: attention refactor and upstream sync apr01 (#365)	10 ماه پیش
AlpinDale	9810daa699 feat: INT8 KV Cache (#298)	11 ماه پیش
AlpinDale	23389d0108 zero out a variable instead of vector in kernels	1 سال پیش
AlpinDale	081545bde6 fix: various CUDA kernel tweaks	1 سال پیش
AlpinDale	05d0a7e763 feat: adapt the attention kernels	1 سال پیش
AlpinDale	3c3944153c feat: add generic attention and FP32 dtype kernels	1 سال پیش