Skip to content

[TRTLLM-14054][perf] Reduce per-step host preparation overhead in the PyTorch executor decode path - #16313

Merged
kaiyux merged 5 commits into
NVIDIA:mainfrom
kaiyux:perf/decode-step-host-prep-20260713
Jul 17, 2026
Merged

[TRTLLM-14054][perf] Reduce per-step host preparation overhead in the PyTorch executor decode path#16313
kaiyux merged 5 commits into
NVIDIA:mainfrom
kaiyux:perf/decode-step-host-prep-20260713

Commits

Commits on Jul 16, 2026