FICHA · MANJARO

tensorflow-opt-cuda

Library for computation using data flow graphs for scalable machine learning (with CUDA and AVX CPU optimizations)

  • Library
  • LIBRARY
  • SCIENCE
  • GPU
  • Dependency only
official+codex · reviewed · May 29, 2026 description in en

Description

Adds a TensorFlow build with both CUDA GPU support and AVX CPU optimizations for systems that can use accelerated machine-learning execution. It is useful for workstations and research machines with compatible NVIDIA GPUs and modern CPUs.

Acceleration depends on matching drivers, CUDA libraries, CPU features, and available memory. Confirm compatibility before installing it on production or shared research systems.

Permissions

Permissions not analysed for this source yet.