Databricks releases, every cloud
- The Genie One MCP server is now generally available
It exposes Genie as a conversational tool over MCP. The previous Beta endpoint is deprecated and will be sunset on October 31, 2026.
- Skills in Genie One support multiple files and folders
Includes up to 3 folders deep and 50 files or folders per skill, limited to text files.
- Account-only user access to Genie One (Beta)
Eligible users are enrolled automatically, governed by Unity Catalog permissions.
- Genie visualizations in Slack replies (Public Preview)
Replaces text-based charts, requires app reinstall for existing setups.
- The counter_diff SQL window function is now available
Calculates differences between consecutive cumulative counter values, available in Databricks Runtime 19 and above.
- Window measures on a numeric index column in metric views
Supports unitless offset and ranges on consecutive integer columns, useful for non-calendar periods like fiscal weeks.
- Run the change, do not just read it. Hands-on labs in your own Databricks workspace, graded when you submit. Browse labs
- Standard performance mode is now available for one-time runs in the Jobs API
Sets performance_target to STANDARD. Supports Apache Airflow's DatabricksSubmitRunOperator.
- Databricks-provided MCP connectors for Genie One and Genie Code are generally available
Supports Google Drive, Gmail, and other tools, governed by Unity Catalog and Unity Gateway.
- Genie One enhancements
Genie Code skills are on by default, can be customized. Supports PowerPoint uploads and backend steering.
- Databricks SQL version 2026.36 is rolling out in Current
It includes features from version 2026.32.
- See which chunks support each ai_search answer (Beta)
Added generate_citations option to show supporting chunks.
- time_bucket SQL function aligns timestamps to fixed-width time buckets
Replaces date_trunc for non-calendar units, requires origin and width parameters.
- Environment version 6 is now available
Supports serverless and standard classic compute.
- Genie One enhancements
Hides Ask Genie button in unsupported embedded dashboards, adds push notifications for scheduled tasks to iOS devices.
- Web search in Genie One (Beta)
Searches public web for current info, citing sources as links.
- Genie One and Genie Agents can now use models served through OpenAI on Databricks
Requires partner-powered AI features to be enabled.
- Genie One web search is in Beta
Searches public web for current info, cites sources as links.
- Databricks SQL version 2026.36 is now available in Preview
It includes features from version 2026.32.
- Get this table by email. One Monday mail covering the week, filtered to the clouds and products you run. Weekly digest
- Genie Code for AI Runtime (Public Preview)
Generates distributed training code and resolves environment issues. Requires AI Runtime.
- Genie One enhancements
Opens Unity Catalog tables directly, adds visualizations to scheduled task emails, and includes a native macOS desktop app.
- The Genie One desktop app for macOS is in Beta
Replaces in-browser access, allows native macOS use.
- Connect to the Databricks Genie app in Microsoft Teams with your workspace URL (Public Preview)
Replaces server hostname with workspace URL for connection.
- Databricks SQL version 2026.32 is rolling out in Current
It is available in the Current channel and includes new features.
- Genie One enhancements
Scheduled tasks maintain consistency by looking up past runs. Genie mobile app allows switching workspaces without signing out.
- Genie One enhancements
Ontology snippets are now publicly available, no request needed. Supports image and Word file uploads.
- Databricks SQL version 2026.32 is now available in Preview
It includes the is_valid_variant function and supports REPLACE USING and REPLACE ON clauses with INSERT INTO.
- REPLACE USING flows for standalone streaming tables (Beta)
Replaces rows matching specified key columns, requires SEQUENCE BY column.
- Alert system tables are in Public Preview
Contains alerts and alert_evaluation_history tables for auditing and monitoring.
- Change data feed on materialized views (Beta)
Requires Databricks Runtime 18 LTS or above. Enables replication to external destinations.
- Databricks-managed MCP connectors are now integrated with Unity Gateway (Beta)
Requires reauthentication to continue using with Genie One and Genie Code.
- Genie One enhancements
Uploads PDF files alongside CSV and Excel. Parses PDFs asynchronously.
- JAR tasks on serverless compute are now generally available
Requires matching Scala, Java Development Kit, and Databricks Connect versions.
- Genie One is now available in Google Sheets
Queries governed data in natural language, importing results as native rows and columns.
- Genie One is now available in Microsoft Excel
Queries governed data in natural language, importing results as native rows and columns.
- Automatic cost attribution for materialized views and streaming tables in Databricks SQL
Inherits custom tags from enclosing SQL warehouse environment for billing.
- Genie One and Genie Agents free usage extended through January 31, 2027
Budget controls do not apply and service principals are excluded.
- Serverless compute access control is now generally available
Governs access to notebooks, jobs, and pipelines through Default Interactive and Automated Compute objects. Existing workloads run unchanged.
- Web terminal on serverless GPU compute (AI Runtime) is in Public Preview
Runs shell commands, monitors GPU usage, and manages files on serverless GPU compute. Requires environment version 5 or above.
- Genie One enhancements
Adds run now option for scheduled tasks and manual workspace URL entry for mobile app sign in.
- Genie One and Genie Agents billed usage corrections
Corrects billed usage to reflect free status through January 31, 2027. Adds correction records to system.billing.usage.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud)
Headlines, dates and product areas are Databricks' own, and every item links to the note it came from. The one-line summaries are ours.