Zero-copy MPI communication of JAX arrays, for turbo-charged HPC applications in Python ⚡

Extending JAX with custom C++ and CUDA code

ALBERT model Pretraining and Fine Tuning using TF2.0

Simple and efficient RevNet-Library for PyTorch with XLA and DeepSpeed support and parameter offload

S + Autograd + XLA :: S-parameter based frequency domain circuit simulations and optimizations using JAX.

基于tensorflow1.x的预训练模型调用，支持单机多卡、梯度累积，XLA加速，混合精度。可灵活训练、验证、预测。

JAX reimplementation and bindings for the NVIDIA cuDecomp library

PyTorch distributed training acceleration framework

Tensorflow2 training code with jit compiling on multi-GPU.

Fast and easy distributed model training examples.

katmer is a powerful library for optimizing the design of optical thin films using automatic differentiation via JAX and Equinox, enabling efficient and accurate inverse design solutions.

Easy to use and blazing fast JAX-based library for high-performance 2D/3D Discrete Element Method (DEM) simulations.

Provides code to serialize the different models involved in Stable Diffusion as SavedModels and to compile them with XLA.

A micro-optimized, high-performance NISQ Statevector Quantum Simulator using JAX XLA Kernel Fusion.

Versatile Data Ingestion Pipelines for Jax

Blueprint for a decentralized, fault-tolerant Surface Code infrastructure leveraging branchless C99 ancilla-syndrome registers, zero-copy C++ binders, and JAX/XLA gradient isolation gates to bypass classical decoding latency walls.

Topological Manifold Control: A multi-paradigm (PyTorch/JAX) geometric morphing engine utilizing differentiable soft-gating, optimized via XLA fused kernels and autograd-isolated non-blocking pipelines.

Classification of multilingual dataset trained only on English training data using pre-trained models. Model is trained on TPUs using PyTorch and torch_xla library.

deep learning inference perf analysis

A Differentiable Data Pipeline Framework for JAX