mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-02-09 20:43:50 +08:00
1c5b0d6a13
* update cubins Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com> * add mtp for fmha_v2 MLA kernels and add chunked-attention support for hopper fmha kernels Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com> --------- Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com> Co-authored-by: Sharan Chetlur <116769508+schetlur-nv@users.noreply.github.com> |
||
|---|---|---|
| .. | ||
| fmha | ||
| convert.cu | ||
| fused_multihead_attention_demo_bert_params.h | ||
| fused_multihead_attention_kernel_1xN_multi_cta.h | ||
| fused_multihead_attention_kernel_1xN_noloop.h | ||
| fused_multihead_attention_kernel_1xN.h | ||
| fused_multihead_attention_kernel_2x2.h | ||
| fused_multihead_attention_kernel_4x1_hopper_noloop.h | ||
| fused_multihead_attention_kernel_4x1_hopper.h | ||
| fused_multihead_attention_kernel_4xN_hopper_noloop.h | ||
| fused_multihead_attention_kernel_4xN_hopper.h | ||
| fused_multihead_attention_kernel.h | ||
| fused_multihead_attention_utils.h | ||
| fused_multihead_attention.cpp | ||
| fused_multihead_attention.h | ||
| fused_multihead_cross_attention_kernel_1xN_noloop.h | ||
| fused_multihead_cross_attention_kernel_1xN.h | ||
| fused_multihead_cross_attention.cpp | ||
| fused_multihead_cross_attention.h | ||
| fused_multihead_flash_attention_kernel_noloop_tiled.h | ||
| fused_multihead_flash_attention_kernel_noloop.h | ||
| fused_multihead_flash_attention_kernel.h | ||
| softmax_bf16.cu | ||
| softmax_fp8.cu | ||
| softmax_fp16.cu | ||
| softmax_fp32.cu | ||
| softmax_impl.h | ||
| softmax_int8.cu | ||