FlashMLA/csrc
chenhongmin.will 7fafcd217d add env
2025-03-01 15:00:46 +08:00
..
cutlass@afa1772203 Initial commit 2025-02-24 09:20:23 +08:00
flash_api.cpp add env 2025-03-01 15:00:46 +08:00
flash_fwd_mla_bf16_sm90.cu update gmem 2025-02-25 09:45:19 +08:00
flash_fwd_mla_fp8_sm90.cu update gmem 2025-02-25 09:45:19 +08:00
flash_fwd_mla_fp16_sm90.cu Merge branch 'main' into will_fp8_mr 2025-02-28 22:15:46 +08:00
flash_fwd_mla_kernel.h enable scale 2025-02-28 20:07:32 +08:00
flash_fwd_mla_metadata.cu support fp16 2025-02-24 01:58:53 -08:00
flash_mla.h update fp8 api 2025-02-26 08:33:25 +08:00
fp8_transpose_v.h update desc 2025-03-01 07:54:39 +08:00
named_barrier.h add transv barrier 2025-02-26 17:57:00 +08:00
softmax.h enable scale 2025-02-28 20:07:32 +08:00
static_switch.h Initial commit 2025-02-24 09:20:23 +08:00
utils.h use mm1's Aregs instead of mma0's Cregs 2025-02-27 11:59:17 +08:00