Skip to content

Sharded Broadcast with Dual-Stream Pipeline for Reduced Peak NPU Memory#257

Open
larksudo wants to merge 6 commits into
LMCache:mainfrom
larksudo:broadcast3
Open

Sharded Broadcast with Dual-Stream Pipeline for Reduced Peak NPU Memory#257
larksudo wants to merge 6 commits into
LMCache:mainfrom
larksudo:broadcast3

fix non-rank0 broadcast oom

2c85d5a
Select commit
Loading
Failed to load commit list.
Sign in for the full log view

Annotations

2 errors and 1 warning
Check code quality
failed Jun 27, 2026 in 43s