david/flash-attention

mirrorاز https://github.com/Dao-AILab/flash-attention

نویسنده	SHA1 پیام	تاریخ
Chirag Jain	50896ec574 Make nvcc threads configurable via environment variable (#885)	9 ماه پیش
Tri Dao	dc08ea1c33 Support H100 for other CUDA extensions	1 سال پیش
Tri Dao	ca81f32e04 Implement rotary embedding in CUDA	2 سال پیش