FICHA · MANJARO

python-tensile

benchmark-driven backend libraries for general matrix-matrix multiplications

  • Library
  • Dev
  • SCIENCE
  • HARDWARE
  • Dependency only
official+codex · reviewed · May 28, 2026 description in en

Description

Provides benchmark-driven backend libraries for general matrix-matrix multiplication, commonly used in GPU and high-performance numerical computing workflows. It helps software choose optimized kernels for linear algebra workloads.

Developers and researchers use it as a performance library, usually through higher-level compute stacks. Results depend on hardware, drivers, generated kernels, and benchmark configuration, so validate performance and numerical behavior on target systems.

Permissions

Permissions not analysed for this source yet.