CPU-first inference orchestration for multi-socket x86. Validates NUMA-aware memory control and heuristic routing can deliver ≥25% throughput uplift on large models without retraining.
-
Updated
May 7, 2026 - Python
#
numa-aware
Here is 1 public repository matching this topic...
Improve this page
Add a description, image, and links to the numa-aware topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the numa-aware topic, visit your repo's landing page and select "manage topics."