Skip to content

llama : disable graph reuse with pipeline parallelism#20463

Merged
ggerganov merged 2 commits into
masterfrom
gg/llama-disable-graph-reuse-with-pp
Mar 12, 2026
Merged

llama : disable graph reuse with pipeline parallelism#20463
ggerganov merged 2 commits into
masterfrom
gg/llama-disable-graph-reuse-with-pp

Revert "CUDA: Improve performance via less synchronizations between …

14be4a4
Select commit
Loading
Failed to load commit list.

Select a check to view from the sidebar