vllm-project/vllm

main

vllm

Grade	Name	Complexity	Churn	Issues
A	benchmarks\kernels\benchmark_cutlass_moe_nvfp4.py	19	1	1
A	benchmarks\kernels\benchmark_device_communicators.py	50	1	-
A	benchmarks\kernels\benchmark_flydsl_moe_w4a16.py	16	1	2
A	benchmarks\kernels\benchmark_fp8_gemm.py	11	1	1
A	benchmarks\kernels\benchmark_fused_collective.py	114	1	2
A	benchmarks\kernels\benchmark_fused_moe_lora_one_shot.py	6	1	-
A	benchmarks\kernels\benchmark_fused_topk.py	7	1	-
A	benchmarks\kernels\benchmark_grouped_gemm_cutlass.py	18	1	1
A	benchmarks\kernels\benchmark_int8_gemm.py	12	1	-
A	benchmarks\kernels\benchmark_layernorm.py	8	1	-
A	benchmarks\kernels\benchmark_lora.py	80	1	-
A	benchmarks\kernels\benchmark_machete.py	46	1	2
A	benchmarks\kernels\benchmark_marlin.py	56	1	2
A	benchmarks\kernels\benchmark_mla_k_concat.py	7	1	-
A	benchmarks\kernels\benchmark_moe.py	119	1	4
A	benchmarks\kernels\benchmark_moe_align_block_size.py	2	1	-
A	benchmarks\kernels\benchmark_moe_defaults.py	13	1	-
A	benchmarks\kernels\benchmark_moe_permute_unpermute.py	22	1	3
A	benchmarks\kernels\benchmark_mrope.py	4	1	-
A	benchmarks\kernels\benchmark_mxfp4_qutlass.py	5	1	2