A transparent, in-container GPU resource controller that enforces memory and compute limits by intercepting CUDA calls without application or driver changes.
-
Updated
Nov 18, 2026 - C
A transparent, in-container GPU resource controller that enforces memory and compute limits by intercepting CUDA calls without application or driver changes.
Hooked CUDA-related dynamic libraries by using automated code generation tools.
To associate your repository with the cuda-hook topic, visit your repo's landing page and select "manage topics."