chore(ci): use new heuristic for throughput benchmarks #2641
gpu_pcc.yml
on: pull_request
Setup instance (cuda-pcc)
2m 0s
Matrix: CUDA post-commit checks
Teardown instance (cuda-pcc)
24s