Repositories list

• maxtext

  Public

  AI-Hypercomputer/maxtext’s past year of commit activity
  A simple, performant and scalable Jax LLM!

  gptmistralfine-tuning+ 11jaxsftlarge-language-modelsllmllama2mixtraldeepseek+ 4

  Python

  •

  Apache License 2.0

  •448•2.1k•75•177•Updated Jan 16, 2026Jan 16, 2026

• pathways-utils

  Public

  AI-Hypercomputer/pathways-utils’s past year of commit activity
  Package of Pathways-on-Cloud utilities

  Python

  •

  Apache License 2.0

  •10•22•0•12•Updated Jan 16, 2026Jan 16, 2026

• accelerator-microbenchmarks

  Public

  AI-Hypercomputer/accelerator-microbenchmarks’s past year of commit activity
  Python

  •

  Apache License 2.0

  •22•9•3•6•Updated Jan 16, 2026Jan 16, 2026

• RecML

  Public
  Python

  •

  Apache License 2.0

  •21•214•3•83•Updated Jan 16, 2026Jan 16, 2026

• gpu-recipes

  Public
  Recipes for reproducing training and serving benchmarks for large machine learning models using GPUs on Google Cloud.

  gpubenchmarksgoogle-cloud-platform+ 2servingdistributed-training

  Python

  •

  Apache License 2.0

  •55•109•6•13•Updated Jan 16, 2026Jan 16, 2026

• maxdiffusion

  Public

  AI-Hypercomputer/maxdiffusion’s past year of commit activity
  Python

  •

  Apache License 2.0

  •53•298•5•20•Updated Jan 15, 2026Jan 15, 2026

• xpk

  Public
  xpk (Accelerated Processing Kit, pronounced x-p-k,) is a software tool to help Cloud developers to orchestrate training jobs on accelerators such as TPUs and GPUs on GKE.

  gkegcloudtpu

  Python

  •

  Apache License 2.0

  •71•159•22•6•Updated Jan 15, 2026Jan 15, 2026

• ml-goodput-measurement

  Public
  Python

  •

  Apache License 2.0

  •6•27•1•0•Updated Jan 15, 2026Jan 15, 2026

• tpu-recipes

  Public
  Shell

  •

  Apache License 2.0

  •43•69•5•12•Updated Jan 14, 2026Jan 14, 2026

• torchprime

  Public
  torchprime is a reference model implementation for PyTorch on TPU.

  Python

  •

  Other

  •13•44•89•15•Updated Jan 6, 2026Jan 6, 2026

• JetStream

  Public
  JetStream is a throughput and memory optimized engine for LLM inference on XLA devices, starting with TPUs (and GPUs in future -- PRs welcome).

  gpuinferencepytorch+ 13transformerllamagptgemmamodel-servingtpujax+ 6

  Python

  •

  Apache License 2.0

  •58•400•14•12•Updated Jan 5, 2026Jan 5, 2026

• google-cloud-mldiagnostics

  Public
  Python

  •

  Apache License 2.0

  •2•6•0•0•Updated Dec 24, 2025Dec 24, 2025

• cloud-accelerator-diagnostics

  Public
  Python

  •

  Apache License 2.0

  •14•26•5•0•Updated Dec 23, 2025Dec 23, 2025

• jetstream-pytorch

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

  inferencepytorchbatching+ 8attentionllamagemmamodel-servingtpullmllm-inference+ 1

  Python

  •

  Apache License 2.0

  •21•80•10•3•Updated Dec 18, 2025Dec 18, 2025

• cloud-diagnostics-xprof

  Public
  Python

  •

  Apache License 2.0

  •4•12•0•0•Updated Dec 12, 2025Dec 12, 2025

• resiliency

  Public
  This project provides resilient training logic to improve the training efficiency of large-scale distributed workloads on Google Cloud.

  Python

  •

  Apache License 2.0

  •2•12•1•0•Updated Sep 29, 2025Sep 29, 2025

• inference-benchmark

  Public
  Python

  •

  Apache License 2.0

  •15•17•7•4•Updated Jun 18, 2025Jun 18, 2025

• kithara

  Public
  Python

  •

  Apache License 2.0

  •8•16•0•2•Updated May 19, 2025May 19, 2025

• ray-tpu

  Public
  Python

  •

  Apache License 2.0

  •7•15•3•1•Updated May 11, 2025May 11, 2025

• aotc

  Public
  Python

  •

  Other

  •3•0•1•0•Updated May 8, 2025May 8, 2025

• cloud-tpu-monitoring-debugging

  Public
  HCL

  •

  Apache License 2.0

  •3•13•0•0•Updated Feb 18, 2025Feb 18, 2025