一键导入
tiny-cutlass
tiny-cutlass 收录了来自 yy6768 的 3 个 skills，并提供仓库级职业覆盖和站内 skill 详情页。
这个仓库中的 skills
Generic workflow and guardrails for developing, verifying, benchmarking, and profiling CUTLASS-based CUDA kernels across any operator family, including GEMM, convolution, attention, fused epilogues, and related experiments. Use when creating or modifying CUTLASS experiments, writing .bat-driven build/verify/bench flows, adding reference tests under csrc/tests, or capturing Nsight Compute CSV and .ncu-rep artifacts.
Profile CUDA kernels with Nsight Compute on SM89 (RTX 4090/4070). Use when the user asks to profile a kernel, analyze its performance, diagnose bottlenecks, read an ncu report, or write an optimization plan — including variants like "profile 一下", "为什么慢", "ncu 报告", "性能分析".
Generic workflow and guardrails for developing, verifying, benchmarking, and profiling CUTLASS-based CUDA kernels across any operator family, including GEMM, convolution, attention, fused epilogues, and related experiments. Use when creating or modifying CUTLASS experiments, writing .bat-driven build/verify/bench flows, adding reference tests under csrc/tests, or capturing Nsight Compute CSV and .ncu-rep artifacts.