Xiaoyu Zhang 3de09aadbc Add new moe wna16 marlin gemm (#14122)

2025-12-01 23:07:53 +08:00

..

gptq

…

marlin

Add new moe wna16 marlin gemm (#14122)

2025-12-01 23:07:53 +08:00

awq_kernel.cu

Fix AWQ Dequant and Weight Loading of deepseek v2 (#6842)

2025-06-17 13:45:10 -07:00

bmm_fp8.cu

move all get_stream in sgl_kernel to c++ to reduce the launch overhead (#12521)

2025-11-02 13:15:05 -08:00

dsv3_fused_a_gemm.cu

…

dsv3_router_gemm_bf16_out.cu

…

dsv3_router_gemm_entry.cu

…

dsv3_router_gemm_float_out.cu

…

fp8_blockwise_gemm_kernel.cu

…

fp8_gemm_kernel.cu

…

int8_gemm_kernel.cu

…

math.hpp

…

nvfp4_expert_quant.cu

…

nvfp4_quant_entry.cu

support cutlass fp4 kernel in sm120 (#11737)

2025-10-29 12:25:16 -07:00

nvfp4_quant_kernels.cu

support cutlass fp4 kernel in sm120 (#11737)

2025-10-29 12:25:16 -07:00

nvfp4_quant.cuh

support cutlass fp4 kernel in sm120 (#11737)

2025-10-29 12:25:16 -07:00

nvfp4_scaled_mm_entry.cu

support cutlass fp4 kernel in sm120 (#11737)

2025-10-29 12:25:16 -07:00

nvfp4_scaled_mm_kernels.cu

support cutlass fp4 kernel in sm120 (#11737)

2025-10-29 12:25:16 -07:00

per_tensor_quant_fp8.cu

…

per_token_group_quant_8bit_v2.cu

…

per_token_group_quant_8bit.cu

…

per_token_quant_fp8.cu

Support hidden_dim % 4 == 0 in per_token_quant_fp8 (#12883)

2025-11-10 17:13:14 +08:00

qserve_w4a8_per_chn_gemm.cu

…

qserve_w4a8_per_group_gemm.cu

…