Kernel Microbenchmark

Build, debug, and interpret vLLM GPU kernel microbenchmarks for CUDA, Triton, and CuteDSL, including CUPTI timing, correctness checks, generated-code inspection, multi-GPU measurements, and SOL sanity checks.

vllm-project Updated

File contents

vllm-project/vllm commit f2dd5be929

Frequently asked questions

npx get-skill add vllm-project/kernel-microbenchmark