Sunny-bot1 6c1f3ff897 topk_gating_softmax support bias (#3405)

2025-08-15 11:57:45 +08:00

..

moe_wna16_marlin_utils

【Inference Optimize】Support automatic generation of marlin kernel (#3149)

2025-08-01 22:43:18 +08:00

deepgemm_preprocess.cu

…

ep_moe_prefill_func.cu

support qk norm (#3145)

2025-08-05 16:46:14 +08:00

fast_hardamard_kernel.cu

optimize w4a8 decoding (#3050)

2025-07-28 22:20:13 +08:00

fast_hardamard_kernel.h

optimize w4a8 decoding (#3050)

2025-07-28 22:20:13 +08:00

fused_moe_helper.h

…

fused_moe_imp_op.h

…

fused_moe_op.h

topk_gating_softmax support bias (#3405)

2025-08-15 11:57:45 +08:00

fused_moe.cu

…

gptq_marlin_repack.cu

…

group_swiglu_with_masked.cu

…

group_swiglu_with_masked.h

…

moe_deepgemm_depermute.cu

…

moe_deepgemm_permute.cu

…

moe_dispatch.cu

…

moe_ffn_wint2.cu

Optimize the performance of moe_expert_ffn_wint2 (#2990)

2025-07-28 10:32:43 +08:00

moe_ffn.cu

optimize w4a8 decoding (#3050)

2025-07-28 22:20:13 +08:00

moe_reduce.cu

…

moe_redundant_topk_select.cu

topk_gating_softmax support bias (#3405)

2025-08-15 11:57:45 +08:00

moe_topk_select.cu

topk_gating_softmax support bias (#3405)

2025-08-15 11:57:45 +08:00

moe_wna16_marlin_gemm.cu

…

moe_wna16_marlin_gemm.h

…

tritonmoe_preprocess.cu

moe preprocess op support 160 experts and fused_moe triton kernel name add K (#3121)

2025-08-01 10:46:20 +08:00

wintx_unzip.cu

…