JAX

2026-11-02

CODING · 1 SOURCE · InfoQ AI, ML & Data Engineering

Google and Meta open-source TorchTPU to provide native PyTorch backend for Google TPUs

Google, in collaboration with Meta, open-sourced TorchTPU, a native PyTorch backend for Google TPUs that prioritizes eager execution while also supporting compiled modes, distributed training, and hardware optimizations; migrating common PyTorch code generally requires only swapping cuda->tpu and nccl->tpu_dist or using torch.accelerator. Google validated TorchTPU across ~5,000 Hugging Face models and reported single-TPUv7 inference of 25.7k token/s (seq_len 5120) and 22.4k token/s (seq_len 10240) for a 30B Muse Glimmer model, plus multi-chip training numbers and large-scale MoE experiments up to 8,192 TPUv7 chips.

8.0

2026-09-30

RESEARCH · 1 SOURCE · arXiv cs.AI

PowerZooJax: JAX-based RL benchmark for power system operation (arXiv:2609.36052v1)

PowerZooJax is an open-source JAX-based benchmark suite for reinforcement learning in power system operation, providing five constrained MDP tasks covering generation, transmission, distribution, distributed energy resources, and a data center microgrid. By rewriting power flow, economic dispatch, market clearing, and device dynamics as JAX computation graphs to keep training and evaluation on GPU, the authors report substantial speedups over CPU simulations and standardized metrics for policy returns, safety violations, and out-of-distribution stress tests.

6.0

2026-09-23

CODING · 1 SOURCE · Hugging Face

NVIDIA demonstrates using Warp and MJWarp to run MuJoCo at GPU scale (up to 2,048 envs)

NVIDIA describes migrating a MuJoCo workflow (an SO-101 follower arm) to MJWarp, an implementation of MuJoCo’s physics pipeline in NVIDIA Warp, and demonstrates advancing as many as 2,048 parallel environments on NVIDIA GPUs. The article explains Warp’s GPU kernel model, device arrays, composable launches and graph capture, and notes Warp features such as differentiable kernels and an opt-in deterministic mode (Warp ≥ 1.15); it focuses on preparing and scaling simulation environments rather than training policies.

6.0

2026-09-23

CODING · 1 SOURCE · InfoQ AI, ML & Data Engineering

Google adds cycle-level kernel profiling to XProf for TPU v7

Google added a Kernel Profiling suite to XProf (the OpenXLA/JAX profiler), exposing cycle-level details for custom kernels (e.g., Pallas) and sampling hardware performance counters on TPU v7 (Ironwood). The tooling shows lowered MLIR, per-unit time-aligned counter tracks, a Perf Counters view with ~16,000 raw counters, and supports periodic sampling (1µs floor) or external-triggered sub-microsecond capture; a Google demo using triple buffering reduced a matmul kernel from 125.5µs to 88µs.

6.0

2026-09-14

MODELS · 1 SOURCE · NVIDIA Developer

Accelerating Dropless MoE Training in JAX with NVIDIA Transformer Engine

NVIDIA Developer describes how to use the NVIDIA Transformer Engine to accelerate dropless Mixture‑of‑Experts (MoE) training workloads in JAX, outlining implementation details and considerations for integrating the engine with MoE models. The article situates this work amid recent MoE models such as DeepSeek, Qwen, and Mixtral and discusses practical steps to improve training efficiency.

6.0

2024-03-28

RESEARCH · 1 SOURCE · Google Research

AutoBNN — compositional Bayesian neural networks for probabilistic time-series forecasting (open-source, Google Research)

Google Research (post by Urs Köster) presents AutoBNN, an open-source JAX package available within TensorFlow Probability that replaces Gaussian processes with compositional Bayesian neural networks to automate discovery of interpretable time-series forecasting models, produce uncertainty estimates, and scale more efficiently to large datasets. AutoBNN maps compositional GP kernels to BNN architectures and supports operators analogous to GP addition and multiplication while enabling GPU/TPU acceleration and possible hybrid architectures with deep BNN components.

7.0

2024-03-06

RESEARCH · 1 SOURCE · Google Research

Croissant 1.0: a schema.org-based metadata format for ML-ready datasets, with support from major dataset hosts and frameworks

Croissant is a new ML-oriented metadata format (v1.0) built on schema.org that standardizes description and organization of datasets without changing underlying file formats. The release includes a spec, example datasets, an open-source Python validator/consumer/generator, a visual editor, and a Responsible AI (RAI) vocabulary extension, and is being adopted by Kaggle, Hugging Face, OpenML, indexed by Dataset Search, and made loadable in major frameworks via TensorFlow Datasets.

7.0