--- title: "Google's Workbench Notebooks Extension Brings Cloud GPUs Into VS Code" subtitle: "Google's new open-source VS Code extension lets you run Workbench Jupyter notebooks on cloud GPUs without leaving your local IDE" url: "https://alphasignal.ai/news/google-s-workbench-notebooks-extension-brings-cloud-gpus-into-vs-code" author: "AlphaSignal Newsroom" author_url: "https://alphasignal.ai/editorial-team" company: "Google for Developers" company_url: "https://alphasignal.ai/company/googledevs" published: "2026-07-01T17:58:21.000Z" updated: "2026-07-02T00:14:12.824Z" type: "News" topics: - "Development" - "Open Source" subtopics: - "Code Generation" upvotes: 212 read_time: "4 min" source_url: "https://www.google.com/" comments: 0 --- # Google's Workbench Notebooks Extension Brings Cloud GPUs Into VS Code > Google's new open-source VS Code extension lets you run Workbench Jupyter notebooks on cloud GPUs without leaving your local IDE If you've ever had to juggle a local VS Code window for writing code and a browser tab running a Jupyter notebook on Google Cloud, Google just closed that loop. The new [Workbench Notebooks VS Code extension](https://marketplace.visualstudio.com/items?itemName=googlecloudtools.workbench-notebooks) lets you connect your local editor directly to a managed Google Cloud Workbench instance and run notebooks on it, no browser required. ## The context-switching problem it solves Until now, these two worlds were mostly separate: developers had customized VS Code environments for project development, and web-based cloud environments for notebook execution, visualization, and training workloads. The result was constant friction -- switching tabs, re-authenticating, losing your IDE keybindings and extensions the moment you needed real compute. The new extension allows you to connect to and run notebooks on managed cloud environments directly within your local IDE, streamlining the ML lifecycle by eliminating context switching and providing direct access to high-performance Google Cloud infrastructure. ## How it actually works Workbench Instances are Jupyter notebook-based development environments for the entire data science workflow. You can interact with other Google Cloud services from within a Workbench instance's Jupyter notebook. Built on top of the existing Jupyter extension, this extension exposes Workbench Jupyter servers directly in VS Code. The kernel selection flow is the key UX detail here. Instead of setting up SSH tunnels or manually configuring remote kernels, the extension surfaces Workbench instances directly in the standard VS Code kernel picker: 1. Open or create a `.ipynb` notebook file in VS Code 2. Click **Select Kernel** > **Google Cloud** > **Workbench** 3. When prompted, sign in, search and select a GCP project, then select an active Workbench instance That's the entire setup. Your notebook cells now execute on the cloud instance, while your editor, extensions, themes, and keybindings stay exactly as they are locally. ## Open source and privacy-forward The newly released extension is fully open-sourced and available on [GitHub](https://github.com/GoogleCloudPlatform/colab-enterprise-vscode) and the VS Code Marketplace. The repo is under an Apache-2.0 license and written primarily in TypeScript, with contributions already open. The extension does not collect any client-side usage data within VS Code. For teams with strict data governance requirements, that's a meaningful detail. To mitigate the risk of malicious extensions accessing your OAuth credentials, Google recommends ensuring the extension is installed from a trusted source and is authored by the verified publisher **GoogleCloudTools**. ## What it's good for This is squarely aimed at ML and data science workflows that need serious compute but don't want to give up a polished local IDE experience. Concrete use cases include: - Training or fine-tuning models on cloud GPUs/TPUs while keeping your full VS Code setup (debugger, Git integration, linting) - Running large-scale data preprocessing on managed infrastructure without provisioning your own VM - Building and training models with simple connectivity to Agent Platform services, including BigQuery, Managed Service for Apache Spark, and Agent Platform integration - Teams that already use Workbench Instances in production and want to stop forcing engineers into a browser-based JupyterLab UI ## What to keep in mind The extension requires an active, running Workbench instance -- it connects to one, it doesn't create one for you. Pricing is based on the compute and services infrastructure you use, as well as other services consumed from within Agent Platform Notebooks. The extension itself is free; you pay for the underlying cloud resources as you normally would. It also requires the Microsoft Jupyter extension as a dependency, so if you're on a locked-down enterprise VS Code setup that restricts extension installs, that's worth checking before rolling this out to a team. The extension can be installed from either the Visual Studio Marketplace or Open VSX , which covers most VS Code-compatible editors including Cursor and Windsurf. ## Part of a broader pattern This release fits into a clear trend of Google moving its cloud notebook infrastructure closer to where developers already work. The Google Cloud Workbench Notebooks extension for VS Code has officially launched, allowing developers to connect their local IDE to scalable, cloud-based Jupyter environments, streamlining the machine learning lifecycle by eliminating context switching and providing direct access to high-performance Google Cloud infrastructure. The open-source angle is also notable. By releasing the extension under Apache-2.0 and inviting community contributions, Google is signaling that this is infrastructure it wants the ecosystem to build on -- not just a closed convenience tool. For teams running heavily customized Workbench setups, the ability to fork and extend the extension to fit internal auth flows or instance-selection logic is a real advantage. ## Sources - [X post by @googledevs](https://x.com/googledevs/status/2072379293435584610) - [GitHub - GoogleCloudPlatform/colab-enterprise-vscode](https://github.com/GoogleCloudPlatform/colab-enterprise-vscode) - [Workbench Notebooks VS Code Extension – Visual Studio Marketplace](https://marketplace.visualstudio.com/items?itemName=GoogleCloudTools.workbench-notebooks) - [Google for Developers Blog – ML Development in VS Code with Google Cloud Power](https://developers.googleblog.com/?hl=en_US) - [What's new with Google Cloud – Strategic Focus](https://strategicfocus.com/2026/05/01/whats-new-with-google-cloud-141/) - [Gemini Enterprise Agent Platform Workbench | Google Cloud](https://cloud.google.com/vertex-ai-notebooks)