[None][perf] Use +64 batch sizes for padding-enabled CUDA graphs#12895

Open

yijingl-nvidia wants to merge 2 commits intoNVIDIA:mainfrom

yijingl-nvidia:default_cuda_graph_batch_sizes

Commits on Apr 3, 2026

[None][perf] Clamp padding-mode CUDA graph batch sizes to max_batch_size
yijingl-nvidia
committed

Commits on Apr 9, 2026

[None][perf] Use sliding-64 batch sizes for padding-enabled CUDA graphs
yijingl-nvidia
committed