Databricks Runtime release notes versions and compatibility
This article lists all Databricks Runtime releases and the schedule for supported releases. Each Databricks Runtime version includes updates that improve the usability, reliability, performance, and security of the Databricks platform.
To learn about the Databricks Runtime support lifecycle, generally available releases, and Beta releases, see Databricks support lifecycles. For information on maintenance updates issued for Databricks Runtime releases, see Databricks Runtime maintenance updates.
Supported Databricks Runtime LTS releases
The following table lists supported Databricks Runtime long-term support (LTS) version releases in addition to the Apache Spark version, release date, and end-of-support date. For optimal lifespan, use a Databricks Runtime LTS version.
LTS means this version is under long-term support. See Databricks Runtime LTS version lifecycle.
Version | Variants | Apache Spark version | Release date | End-of-support date |
|---|---|---|---|---|
16.4 LTS | 3.5.2 | May 9, 2025 | May 9, 2028 | |
15.4 LTS | 3.5.0 | Aug 19, 2024 | Aug 19, 2027 | |
14.3 LTS | 3.5.0 | Feb 1, 2024 | Feb 1, 2027 | |
13.3 LTS | 3.4.1 | Aug 22, 2023 | Aug 22, 2026 | |
12.2 LTS | 3.3.2 | Mar 1, 2023 | Mar 1, 2026 | |
11.3 LTS | 3.3.0 | Oct 19, 2022 | Oct 19, 2025 |
All supported Databricks Runtime releases
The following table lists all the Apache Spark version, release date, and end-of-support date for supported Databricks Runtime releases. For optimal lifespan, use a Databricks Runtime LTS version.
Version | Variants | Apache Spark version | Release date | End-of-support date |
|---|---|---|---|---|
17.1 (Beta) | 4.0.0 | Jul 15, 2025 | Jan 23, 2026 | |
17.0 | 4.0.0 | Jun 24, 2025 | Nov 20, 2025 | |
16.4 LTS | 3.5.2 | May 9, 2025 | May 9, 2028 | |
16.3 | 3.5.0 | Mar 31, 2025 | Sep 1, 2025 | |
16.2 | 3.5.0 | Feb 5, 2025 | Aug 5, 2025 | |
15.4 LTS | 3.5.0 | Aug 19, 2024 | Aug 19, 2027 | |
14.3 LTS | 3.5.0 | Feb 1, 2024 | Feb 1, 2027 | |
13.3 LTS | 3.4.1 | Aug 22, 2023 | Aug 22, 2026 | |
12.2 LTS | 3.3.2 | Mar 1, 2023 | Mar 1, 2026 | |
11.3 LTS | 3.3.0 | Oct 19, 2022 | Oct 19, 2025 |
MLflow-Databricks Runtime compatibility matrix
This section lists Databricks Runtime ML versions and their respective MLflow versions.
Databricks Runtime ML version | MLflow version |
|---|---|
17.1 (Beta) | |
17.0 | |
16.4 LTS | |
16.3 | |
16.2 | |
15.4 LTS | |
14.3 LTS | |
13.3 LTS - 14.0 | |
12.2 LTS | |
11.3 LTS |
Feature Engineering compatibility matrix
This section lists Databricks Runtime ML versions and their respective Feature Engineering and Workspace Feature Store client versions.
Databricks Runtime ML version |
|
|
|---|---|---|
17.1 (Beta) | 0.12.1 | None |
17.0 | 0.10.2 | None |
16.4 LTS | 0.10.2 | None |
16.3 | 0.9.0 | None |
16.2 | 0.8.0 | None |
15.4 LTS | 0.6.x* | None |
14.3 LTS | 0.2.x | None |
13.3 LTS | 0.1.x | 0.14.1 |
12.2 LTS | Not supported | 0.10.0 |
11.3 LTS | Not supported | 0.7.0 (requires MLflow < 2.0) |
- For compute created on or after March 31, 2025 that is not enabled for Photon, the installed version of
databricks-feature-engineeringis 0.8.0.
Apache Spark migration guidance
Find Spark-specific migration information in the Apache Spark documentation. The migration information for each Spark version can be found at a URL like the following:
https://spark.apache.org/docs/<version>/migration-guide.html.
Replace <version> with the Spark version included in the Databricks Runtime version you're migrating to. For example, the URL with migration information for Spark 3.5.0, included in Databricks Runtime 14.3 LTS, is https://spark.apache.org/docs/3.5.0/migration-guide.html.
Beta releases
Version | Variants | Apache Spark version | Beta release date |
|---|---|---|---|
17.1 (Beta) | 4.0.0 | Jul 15, 2025 | |
17.1 ML (Beta) | 4.0.0 | Jul 15, 2025 |
Unsupported releases
For information on unsupported Databricks Runtime version release notes, see End-of-support Databricks Runtime release notes. The unsupported Databricks Runtime versions have been retired and might not be updated.