Serverless environment versions
This page provides a reference for the currently released serverless environment versions.
Serverless compute for notebooks and jobs uses environment versions, which provide a stable client API based on Spark Connect to ensure application compatibility. This allows Databricks to upgrade the server independently, delivering performance improvements, security enhancements, and bug fixes without requiring any code changes to workloads.
Each environment version includes a specific Python version and a set of Python packages with defined versions. Databricks introduces new features and fixes in the latest environment version while applying security updates to all supported environment versions.
Environment version support dates
To provide a consistent and predictable experience, Databricks supports a three-year lifecycle for all serverless environment versions.
Serverless GPU compute is in Beta.
Serverless GPU environment versions are built on top of the corresponding CPU environment versions. The serverless GPU environments include additional packages to facilitate single and multi-node deep learning workloads.
Version | Variants | System environment | Release date | End-of-support date |
|---|---|---|---|---|
4 |
| August 13, 2025 | August 13, 2028 | |
3 |
| June 13, 2025 | June 13, 2028 | |
2 |
| November 21, 2024 | November 21, 2027 | |
1 |
| March 12, 2024 | March 12, 2027 |
Environment version end of support
When an environment version reaches its end-of-support date:
- The environment version will no longer be selectable for new jobs and notebooks in the UI, but will be configurable for jobs using the API.
- Existing jobs and notebooks running on the environment version will continue to run.
- You will be able to import and run notebooks that reference the environment version.
- Workloads running on the environment version might experience longer start-up times.
- Creating new Unity Catalog UDFs that reference the environment version will trigger a warning, but will still run.