Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

JaxPP is a library for JAX that enables flexible MPMD pipeline parallelism for large-scale LLM training

MLIR-based partitioning system

Powering AWS purpose-built machine learning chips. Blazing fast and cost effective, natively integrated into PyTorch and TensorFlow and integrated with your favorite AWS services

torchprime is a reference model implementation for PyTorch on TPU.

A machine learning compiler for GPUs, CPUs, and ML accelerators

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Enabling PyTorch on XLA Devices (e.g. Google TPU)

PyTorch/XLA integration with JetStream (https://github.com/google/JetStream) for LLM inference"

Enabling PyTorch on XLA Devices (e.g. Google TPU)