kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
6df2c8a074
TensorRT-LLMs/cpp/kernels/fmha_v2/src
History
Bo Li a66eeab537
[TRTLLM-9805][feat] Skip Softmax Attention. (#9821) 
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
Signed-off-by: Tian Zheng <29906817+Tom-Zheng@users.noreply.github.com>
Co-authored-by: Tian Zheng <29906817+Tom-Zheng@users.noreply.github.com>
2025-12-21 02:52:42 -05:00
..
fmha [TRTLLM-9805][feat] Skip Softmax Attention. (#9821) 2025-12-21 02:52:42 -05:00
convert.cu [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_demo_bert_params.h [TRTLLM-9805][feat] Skip Softmax Attention. (#9821) 2025-12-21 02:52:42 -05:00
fused_multihead_attention_kernel_1xN_multi_cta.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel_1xN_noloop.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel_1xN.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel_2x2.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel_4x1_hopper_noloop.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel_4x1_hopper.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel_4xN_hopper_noloop.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel_4xN_hopper.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_kernel.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention_utils.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_attention.cpp [TRTLLM-9805][feat] Skip Softmax Attention. (#9821) 2025-12-21 02:52:42 -05:00
fused_multihead_attention.h [TRTLLM-9805][feat] Skip Softmax Attention. (#9821) 2025-12-21 02:52:42 -05:00
fused_multihead_cross_attention_kernel_1xN_noloop.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_cross_attention_kernel_1xN.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_cross_attention.cpp [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_cross_attention.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_flash_attention_kernel_noloop_tiled.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_flash_attention_kernel_noloop.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
fused_multihead_flash_attention_kernel.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
softmax_bf16.cu [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
softmax_fp8.cu [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
softmax_fp16.cu [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
softmax_fp32.cu [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
softmax_impl.h [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00
softmax_int8.cu [#8476][chore] Update license (#8807) 2025-11-19 15:05:25 -08:00