DLT release 2025.12

March 27 - April 2, 2025

These features and improvements were released with the 2025.12 release of DLT.

Databricks Runtime versions used by this release

Channel:

• CURRENT (default): Databricks Runtime 15.4
• PREVIEW: Databricks Runtime 15.4 or 16.1

note

Because DLT channel releases follow a rolling upgrade process, channel upgrades are deployed to different regions at different times. Your release, including Databricks Runtime versions, might not be updated until a week or more after the initial release date. To find the Databricks Runtime version for a pipeline, see Runtime information.

New features and improvements

• CREATE PRIVATE { MATERIALIZED VIEW | STREAMING TABLE } is now supported in Lakeflow Declarative Pipelines. Use the PRIVATE parameter to create materialized views or streaming tables without publishing them to a catalog. To learn more, see CREATE STREAMING TABLE (Lakeflow Declarative Pipelines) and CREATE MATERIALIZED VIEW (Lakeflow Declarative Pipelines).
• You can now use CREATE TEMPORARY VIEW in Lakeflow Declarative Pipelines to create a view that is only persisted within the pipeline. To learn more, see CREATE VIEW (Lakeflow Declarative Pipelines).
• Improved the INITIALIZING performance for pipeline updates.

Bug fixes

• This release fixes an uncommon issue where Hive metastore streaming checkpoints are stored and read from the DBFS root instead of the location specified by the user.
• This release fixes an issue where an event hook fails to make forward progress.