Lumina 81b16a2bc9 [Kernel] Better inf handling for grouped topk cu (#24886) ...

Signed-off-by: lumina37 <starry.qvq@gmail.com>

2025-09-18 05:53:55 +00:00

..

marlin_moe_wna16

…

permute_unpermute_kernels

Fix CUDA permute/unpermute for use with DeepGemm Moe (#17934)

2025-07-27 07:08:00 -07:00

grouped_topk_kernels.cu

[Kernel] Better inf handling for grouped topk cu (#24886)

2025-09-18 05:53:55 +00:00

moe_align_sum_kernels.cu

[perf] Speed up align sum kernels (#21079)

2025-07-21 11:19:23 -07:00

moe_ops.h

[Kernel] Add fused grouped_topk kernel for MoE (#23274)

2025-08-25 11:47:52 -07:00

moe_permute_unpermute_op.cu

…

moe_wna16_utils.h

pre-commit autoupdate (#17380)

2025-04-29 06:46:55 -07:00

moe_wna16.cu

…

topk_softmax_kernels.cu

Apply fixes for CUDA 13 (#24599)

2025-09-17 09:15:42 -04:00

torch_bindings.cpp

…