courage17340 981cadb35c [Bugfix][Kernel] fix merge attn states when both prefix and suffix are empty (#28181) ...

Signed-off-by: courage17340 <courage17340@163.com>

2025-11-06 17:52:13 +08:00

..

mla

[Attention] Tune CUTLASS MLA num_splits (#26846)

2025-10-16 06:36:09 -07:00

attention_dtypes.h

…

attention_generic.cuh

…

attention_kernels.cuh

[Refactor] Refactor FP8 & INT8 Quant Folder inside w8a8 (#25293)

2025-10-08 10:20:48 -04:00

attention_utils.cuh

…

dtype_bfloat16.cuh

…

dtype_float16.cuh

…

dtype_float32.cuh

…

dtype_fp8.cuh

…

merge_attn_states.cu

[Bugfix][Kernel] fix merge attn states when both prefix and suffix are empty (#28181)

2025-11-06 17:52:13 +08:00

paged_attention_v1.cu

[Bugfix][ROCm] Fix for warp_size uses on host (#21205)

2025-07-24 00:37:19 -07:00

paged_attention_v2.cu

[Bugfix][ROCm] Fix for warp_size uses on host (#21205)

2025-07-24 00:37:19 -07:00

vertical_slash_index.cu

Implements dual-chunk-flash-attn backend for dual chunk attention with sparse attention support (#11844)

2025-05-12 19:52:47 -07:00