JAX: High-Performance Array Computing#

JAX is Autograd and XLA, brought together for high-performance numerical computing.

Familiar API

JAX provides a familiar NumPy-style API for ease of adoption by researchers and engineers.

Transformations

JAX includes composable function transformations for compilation, batching, automatic differentiation, and parallelization.

Run Anywhere

The same code executes on multiple backends, including CPU, GPU, & TPU

Getting Started
User Guides
Developer Docs