triple-mu 444013585d Fix typos and unify size(s)/stride(s) API calls (#8799)

2025-08-08 00:18:08 -07:00

..

awq_kernel.cu

…

bmm_fp8.cu

…

dsv3_fused_a_gemm.cu

Fix typos and unify size(s)/stride(s) API calls (#8799)

2025-08-08 00:18:08 -07:00

dsv3_router_gemm_bf16_out.cu

[Kimi K2] dsv3_router_gemm supports NUM_EXPERTS == 384 (#8013)

2025-08-01 22:01:24 +08:00

dsv3_router_gemm_entry.cu

[Kimi K2] dsv3_router_gemm supports NUM_EXPERTS == 384 (#8013)

2025-08-01 22:01:24 +08:00

dsv3_router_gemm_float_out.cu

[Kimi K2] dsv3_router_gemm supports NUM_EXPERTS == 384 (#8013)

2025-08-01 22:01:24 +08:00

fp8_blockwise_gemm_kernel.cu

Fix typos and unify size(s)/stride(s) API calls (#8799)

2025-08-08 00:18:08 -07:00

fp8_gemm_kernel.cu

Fix typos and unify size(s)/stride(s) API calls (#8799)

2025-08-08 00:18:08 -07:00

int8_gemm_kernel.cu

Fix typos and unify size(s)/stride(s) API calls (#8799)

2025-08-08 00:18:08 -07:00

nvfp4_expert_quant.cu

…

nvfp4_quant_entry.cu

…

nvfp4_quant_kernels.cu

…

nvfp4_scaled_mm_entry.cu

…

nvfp4_scaled_mm_kernels.cu

Fix typos and unify size(s)/stride(s) API calls (#8799)

2025-08-08 00:18:08 -07:00

per_tensor_quant_fp8.cu

[sgl-kernel performace] fix fp8 quant kernels dispatch __nv_fp8_e4m3 bug to improve performance 10%-20% (#8499)

2025-07-29 23:31:54 +08:00

per_token_group_quant_8bit.cu

[sgl-kernel performace] fix fp8 quant kernels dispatch __nv_fp8_e4m3 bug to improve performance 10%-20% (#8499)

2025-07-29 23:31:54 +08:00

per_token_quant_fp8.cu

[sgl-kernel] avoid per_token_quant_fp8.cu hardcode sm_count (#8738)

2025-08-04 12:55:57 +08:00

qserve_w4a8_per_chn_gemm.cu

…

qserve_w4a8_per_group_gemm.cu

…