Databricks releases, every cloud
- Login required to download ODBC driver
Requires login and license acceptance, except for AWS GovCloud which needs account team access.
- New resource types for Databricks Apps
Adds MLflow experiments, vector search indexes, UDFs, and Unity Catalog connections as resources.
- Serverless compute is now available in Azure China North 3
Available for notebooks, jobs, pipelines, and SQL warehouses.
- Confluence connector (Beta)
Ingests Confluence spaces and pages into Databricks, requires OAuth U2M configuration.
- PostgreSQL connector in Lakeflow Connect (Public Preview)
Enables incremental data ingestion from various PostgreSQL databases. Requires configuration.
- Delta Sharing to external Iceberg clients is in Public Preview
Supports Snowflake, Trino, Flink, and Spark clients with zero-copy access.
- Run the change, do not just read it. Hands-on labs in your own Databricks workspace, graded when you submit. Browse labs
- Serverless workspaces are now available in new GCP regions
Available in most GCP regions, see documentation for list.
- Lakebase now in Public Preview
Introduces autoscaling compute and database branching on AWS. Free until January 2026.
- Dashboard visualization enhancements and fixes
Sorts charts by hidden measure fields, adds spacing to range filters, and fixes Safari cross-filtering.
- Disable legacy features settings are now GA
Disables access to DBFS, Hive Metastore, and shared compute in new workspaces.
- Genie enhancements and fixes
Shows Service Principal ID in Monitoring page, fixes API suggested questions issue, and updates Research Agent architecture.
- OpenAI GPT-5.2 now available as a Databricks-hosted model
Accessible via Foundation Model APIs on a pay-per-token basis.
- Change owner for materialized views or streaming tables defined in Databricks SQL
Done through Catalog Explorer, replacing previous methods.
- Customizable SharePoint connector (Beta)
Ingests structured, semi-structured, and unstructured files into Delta tables with customizable schema inference and parsing options.
- Discover files in Auto Loader efficiently using file events
Requires file events enabled on external locations. Replaces directory listing for efficiency.
- Jira connector (Beta)
Ingests Jira issues, comments, and attachments metadata into Databricks.
- Microsoft Dynamics 365 connector (Public Preview)
Ingests data from Dynamics 365 apps into Databricks. Requires Lakeflow Connect.
- NetSuite connector (Public Preview)
Ingests data from NetSuite2.com via API, CLI, or notebook.
- Get this table by email. One Monday mail covering the week, filtered to the clouds and products you run. Weekly digest
- Databricks Runtime 18.0 and Databricks Runtime 18.0 ML are in Beta
Powered by Apache Spark 4.1.0, includes JDK 21 as default.
- Databricks Runtime maintenance updates (12/09)
Includes bug fixes, security patches, and performance improvements for versions 12.2 to 17.3.
- ForEachBatch for Spark Declarative Pipelines on Lakeflow is available (Public Preview)
Allows processing streams as micro-batches in Python. Requires Lakeflow setup.
- New columns in Lakeflow system tables (Public Preview)
New columns provide job monitoring and troubleshooting data. They are not populated for rows before December 2025.
- Expanded regional availability for C5 and TISAX compliance
Now available in all regions and with serverless compute.
- New token expiration policy for open Delta Sharing
New tokens expire in 1 year, existing ones expire on December 8, 2026.
- Vector Search reranker is now generally available
Improves retrieval quality, see documentation for usage.
- Dashboard visualization enhancements and fixes
Adds individual series labels and fixes download format options to include CSV, TSV, and Excel.
- Genie Research Agent, benchmark, and answer improvements
Genie now asks clarifying questions, explains benchmark errors, and collapses query result tables when summaries are sufficient.
- Built-in Excel file format support (Beta)
Supports reading Excel files via Spark DataFrames without external libraries.
- Dashboard visualization enhancements
Pivot table headers render faster with reduced scroll lag.
- Genie Agent (formerly Genie Space) limits and Research Agent improvements
Limits increased to 200, adds starter questions and auto-suggested benchmarks.
- Spark Declarative Pipelines on Lakeflow stream progress metrics are generally available
Supports querying event log for stream progress metrics.
- Anthropic Claude Opus 4.5 now available as a Databricks-hosted model
Accessible via Foundation Model APIs and AI Functions, supports pay-per-token and batch inference workloads.
- Automate classic workspace creation with AWS IAM temporary delegation
Replaces AWS Quickstart template, grants temporary access to provision IAM roles and S3 bucket.
- Configure full refresh behavior for managed database connectors
Applies to Lakeflow Connect managed database connectors, allowing scheduled full refresh snapshots and automatic recovery from unsupported schema changes.
- Real-time collaboration in notebook cells, files, and SQL editor
Multiple users can edit the same cell simultaneously, viewing each other's edits.
- Serverless compute is now available in new regions
Added to Australia Southeast, Canada Central, Mexico Central, Norway East, and South Africa North regions.
- Use Git CLI commands in Git folders (Beta)
Runs standard Git commands in Databricks web terminal. Requires Git folder setup.
- Automatic email notifications for expiring personal access tokens (Public Preview)
Sends emails 7 days before token expiration.
- Configure compute size for Databricks Apps (Public Preview)
Controls CPU, memory, and cost based on workload requirements.
- Dashboard visualization improvements and bug fixes
Counters now track metrics over time with sparklines. Filters apply instantly by default.
Headlines, dates and product areas are Databricks' own, and every item links to the note it came from. The one-line summaries are ours.