Skip to content

fix(inference): materialise checkpoint tensors in host memory before H2D copy (17.7x faster load on GB300) - #150

Open
rwagwani wants to merge 2 commits into
NVIDIA:mainfrom
rwagwani:fix/checkpoint-load-mmap-h2d
Open

fix(inference): materialise checkpoint tensors in host memory before H2D copy (17.7x faster load on GB300)#150
rwagwani wants to merge 2 commits into
NVIDIA:mainfrom
rwagwani:fix/checkpoint-load-mmap-h2d

fix(inference): apply the host-memory staging copy conditionally

3d8c469
Select commit
Loading
Failed to load commit list.
Sign in for the full log view
pre-commit
succeeded Aug 5, 2026 in 57s