FICHA · AUR

python-jax

Composable transformations of Python+NumPy programs

  • JAX numerical computing library
  • LIBRARY
  • AI
  • SCIENCE
  • Dependency only
official+codex · reviewed · Jun 3, 2026 description in en

Description

NumPy-style numerical programs can be transformed for automatic differentiation, vectorization, and accelerator execution. ML engineers and researchers use JAX for training, simulation, and high-performance array workloads. Datasets, model code, and accelerator usage need validation and privacy review.

Permissions

Permissions not analysed for this source yet.