Introduction to Colab Enterprise

Colab Enterprise is a collaborative, managed notebook environment with the security and compliance capabilities of Google Cloud.

Share and collaborate

Colab Enterprise lets you share notebooks (IPYNB files) and collaborate with others. You can share a notebook with a single user, Google group, or Google Workspace domain. You control this access through Identity and Access Management (IAM).

Learn how to grant access to a notebook.

Managed compute

Colab Enterprise lets you work in notebooks without having to manage infrastructure. Colab Enterprise provisions a runtime for you when you need it. If you want to, you can configure runtimes for specific needs, but Colab Enterprise starts them for you and shuts them down when you no longer need them.

Configurable runtimes

Use the default runtime or configure your own by creating a runtime template. When you create a runtime template, you can choose between any available machine type, add accelerators, and specify your disk space.

By using runtimes that you configured, you can run code on the runtime that meets your needs at the moment. For example, you can test your code with a small dataset by using a runtime with a smaller, cost-effective machine type. Then you can use a runtime with more processing power when you need to run your code with more data.

To learn more, see Runtimes and runtime templates.

Integrated with Vertex AI and BigQuery

Colab Enterprise's integrations with Google Cloud services make it easier to use notebooks that interact with those services.

Colab Enterprise is a part of Vertex AI. Colab Enterprise's API methods are a subset of the Vertex AI API. This can save you time when you're managing access to Colab Enterprise resources or when you authenticate and authorize your notebook code to use Vertex AI.

To learn more, see the API usage overview.

Colab Enterprise is embedded within the UI of its integrated services. This lets you do work in Vertex AI and BigQuery, and then continue your work in a notebook without leaving their areas of the Google Cloud console. For example, you can create a notebook directly from the BigQuery editor, populated with a sample query to help you get started. To learn more, see BigQuery's Introduction to notebooks.

Inline code completion

Colab Enterprise lets you write code with Gemini assistance. Use the inline code completion suggestions to complete coding projects faster. To learn more, see write code with Gemini assistance.

Idle shutdown

To help manage costs, Colab Enterprise runtimes shut down when they're inactive for a specific time period. You can change the amount of time or turn this feature off.

For more information, see Idle shutdown.

Run code that interacts with Google Cloud services and APIs

By default Colab Enterprise notebooks use your user credentials to authenticate and authorize code that interacts with other Google Cloud services.

This means that the notebook's code has the same level of access to Google Cloud that the user does. This makes it easier to write and run code that interacts with Google Cloud services.

For more information, see Run code that interacts with Google Cloud.

Colab Enterprise is different from Colab

Colab Enterprise is different from Colaboratory, which is still available. Both have specific advantages that can be helpful depending on your needs. See the main differences in the following table:

Component Colab Enterprise Colab
Storage Regional storage in Dataform Google Drive storage, regionalized storage not supported
Access control Managed by IAM Managed through Google Drive sharing functionality
Security and networking Google Cloud security and network functionality Google Drive-based security, with the internet always available
Support Google Cloud support Send feedback or report a bug

Limitations

Consider the following limitations of Colab Enterprise when planning your project:

  • The file size of uploaded notebooks is limited to approximately 20 MB. Working with a notebook that is near the size limit can impact performance.

  • Shielded VM runtimes can't be upgraded.

See also Quotas and limits.

What's next