Skip to main content
Google Cloud
Documentation Technology areas
  • AI and ML
  • Application development
  • Application hosting
  • Compute
  • Data analytics and pipelines
  • Databases
  • Distributed, hybrid, and multicloud
  • Generative AI
  • Industry solutions
  • Networking
  • Observability and monitoring
  • Security
  • Storage
Cross-product tools
  • Access and resources management
  • Costs and usage management
  • Google Cloud SDK, languages, frameworks, and tools
  • Infrastructure as code
  • Migration
Related sites
  • Google Cloud Home
  • Free Trial and Free Tier
  • Architecture Center
  • Blog
  • Contact Sales
  • Google Cloud Developer Center
  • Google Developer Center
  • Google Cloud Marketplace
  • Google Cloud Marketplace Documentation
  • Google Cloud Skills Boost
  • Google Cloud Solution Center
  • Google Cloud Support
  • Google Cloud Tech Youtube Channel
/
  • English
  • Deutsch
  • Español
  • Español – América Latina
  • Français
  • Indonesia
  • Italiano
  • Português
  • Português – Brasil
  • 中文 – 简体
  • 中文 – 繁體
  • 日本語
  • 한국어
Console Sign in
  • Cloud TPU
Guides Reference Support Resources
Contact Us Start free
Google Cloud
  • Documentation
    • Guides
    • Reference
    • Support
    • Resources
  • Technology areas
    • More
  • Cross-product tools
    • More
  • Related sites
    • More
  • Console
  • Contact Us
  • Start free
  • Discover
  • Introduction to Cloud TPU
  • TPU architecture
  • TPU software versions
  • TPU versions
    • TPU v6e
    • TPU v5p
    • TPU v5e
    • TPU v4
    • TPU v3
    • TPU v2
  • Regions and zones
  • TPU consumption options
  • Get started
  • All getting started guides
  • Set up the Cloud TPU environment
  • Reserve TPUs
    • About TPU reservations
    • Request a short-term reservation with calendar mode
    • Request a long-term reservation
    • Share a reservation
    • Consume a reservation
  • Run JAX on Cloud TPU VM
  • Run PyTorch on Cloud TPU VM
  • Train on Cloud TPU slices
  • Run JAX on Cloud TPU slices
  • Run PyTorch on Cloud TPU slices
  • Configure TPUs
  • Connect a TPU to a shared VPC network
  • Connect to a TPU VM without a public IP address
  • Configure networking and access
  • Use a cross-project service account
  • Storage options
    • Storage options for Cloud TPU
    • Attach durable block storage to a TPU VM
    • Connect to Cloud Storage buckets
    • Mount a Filestore instance on a TPU VM
  • Training and inference
  • Trillium (v6e) introduction
  • Train a model using v5e
  • TPU inference
  • Multislice training
  • Scale ML workloads using Ray
  • Run TPU applications in a Docker container
  • TPUs in GKE introduction
  • Work with image datasets
    • Convert an image classification dataset for use with Cloud TPU
    • Download, pre-process and upload the ImageNet dataset
    • Download, pre-process and upload the COCO dataset
  • Manage TPUs
  • Manage TPU resources
  • Manage queued resources
  • Request TPUs using Flex-start
  • Manage TPU Spot VMs
  • Prepare for maintenance events
  • Schedule TPU collections for inference workloads
  • Autocheckpoint
  • View maintenance notifications
  • Manually start host maintenance
  • Preemptible TPUs
  • Optimize performance
  • Cloud TPU performance guide
  • Improve your model's performance with bfloat16
  • Monitor and troubleshoot TPUs
  • Troubleshoot TPU VMs
  • Monitor TPU VMs
  • Dashboards for monitoring and logging
  • Troubleshoot TensorFlow models
  • Troubleshoot PyTorch models
  • Troubleshoot JAX models
  • Cloud TPU error glossary
  • Cloud TPU audit logs
  • Profile TPUs
  • Profile TPU VMs
  • Profile Multislice environments
  • Profile PyTorch XLA workloads
  • Tutorials
  • All tutorials
  • ResNet
  • Serve generative AI models
    • JetStream MaxText inference on v6e
    • JetStream PyTorch inference on v6e
    • MaxDiffusion inference on v6e
    • vLLM inference on v6e
    • Serve an LLM with vLLM (GKE)
  • Interactive notebooks
  • All interactive notebooks
  • HuggingFace Diffusers with PyTorch/XLA 2.0
  • Distributed PyTorch/XLA Basics
  • AI and ML
  • Application development
  • Application hosting
  • Compute
  • Data analytics and pipelines
  • Databases
  • Distributed, hybrid, and multicloud
  • Generative AI
  • Industry solutions
  • Networking
  • Observability and monitoring
  • Security
  • Storage
  • Access and resources management
  • Costs and usage management
  • Google Cloud SDK, languages, frameworks, and tools
  • Infrastructure as code
  • Migration
  • Google Cloud Home
  • Free Trial and Free Tier
  • Architecture Center
  • Blog
  • Contact Sales
  • Google Cloud Developer Center
  • Google Developer Center
  • Google Cloud Marketplace
  • Google Cloud Marketplace Documentation
  • Google Cloud Skills Boost
  • Google Cloud Solution Center
  • Google Cloud Support
  • Google Cloud Tech Youtube Channel
  • Home
  • Cloud TPU
  • Documentation
  • Guides

All interactive notebooks
Stay organized with collections Save and categorize content based on your preferences.

PyTorch Kaggle notebooks

  • PyTorch/XLA 2.0 on Kaggle

    Use PyTorch/XLA 2.0 on the Kaggle TPU VM accelerator to run a Stable Diffusion image example.

  • Distributed PyTorch/XLA Basics

    Perform distributed operations on a TPU VM using PyTorch/XLA with the PJRT runtime.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

  • Why Google

    • Choosing Google Cloud
    • Trust and security
    • Modern Infrastructure Cloud
    • Multicloud
    • Global infrastructure
    • Customers and case studies
    • Analyst reports
    • Whitepapers
  • Products and pricing

    • See all products
    • See all solutions
    • Google Cloud for Startups
    • Google Cloud Marketplace
    • Google Cloud pricing
    • Contact sales
  • Support

    • Google Cloud Community
    • Support
    • Release Notes
    • System status
  • Resources

    • GitHub
    • Getting Started with Google Cloud
    • Google Cloud documentation
    • Code samples
    • Cloud Architecture Center
    • Training and Certification
    • Developer Center
  • Engage

    • Blog
    • Events
    • X (Twitter)
    • Google Cloud on YouTube
    • Google Cloud Tech on YouTube
    • Become a Partner
    • Google Cloud Affiliate Program
    • Press Corner
  • About Google
  • Privacy
  • Site terms
  • Google Cloud terms
  • Manage cookies
  • Our third decade of climate action: join us
  • Sign up for the Google Cloud newsletter Subscribe
  • English
  • Deutsch
  • Español
  • Español – América Latina
  • Français
  • Indonesia
  • Italiano
  • Português
  • Português – Brasil
  • 中文 – 简体
  • 中文 – 繁體
  • 日本語
  • 한국어