Releasesby GlyphDatabricks Runtime 19: September 21, 2026
New features and improvements truncate function : SQL and the Scala and Python DataFrame APIs now support truncate , which truncates a number to a specified scale without rounding.
Also shipped Sep 21, 2026Databricks in September 2026 →
BigQuery Studio query results pane shows recent run history
The Query results pane now shows a short history of recent runs for a query, including multi-statement queries, without switching to the Job history tab. It is generally available.
Cortex AI Function Evaluation for measuring quality (Public Preview)
Snowflake Cortex AI Function Evaluation is now available in public preview, enabling customers to measure the output quality, cost, and token usage of custom AI Functions and Cortex AI calls against labeled datasets.
Cortex AI Function Optimization for more efficient AI implementations (Public Preview)
Snowflake Cortex AI Function Optimization is now available in public preview, enabling customers to automatically search across prompts and models for improved implementations of custom AI Functions.
Cortex Agents Compact API (Preview)
The Cortex Agents Compact API is now available in preview. Use the agent:compact endpoint to summarize a conversation and pass the compact representation to subsequent agent:run requests. Compacting conversation history reduces token consumption and helps conversations stay within the model context window.
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes. This feature is generally available ( GA ). It accelerates vector search performance and increases queries per second (QPS) for vector workloads. For more information, see Accelerate queries with the Columnar Engine .
SWE-2 Research Preview in the Agent Selector
SWE-2, Cognition's next-generation software engineering model, is rolling out as a research preview in the agent selector under the "Research preview" section. Choose SWE-2 and a reasoning effort (Medium, High, or Max) when starting a session, or switch to it mid-session with the agent toggle next to the message input; in Slack, use !swe2 .
More from Databricks Data Intelligence PlatformEvery Databricks Data Intelligence Platform release →
The query history system table is now generally available
The system.query.history system table is now generally available. Use it to access account-wide records for queries run on SQL warehouses or serverless compute, including execution status, duration metrics, and query source. See Query history system table reference .
Bundled BigQuery JDBC driver removal in Databricks Runtime 20
Starting with Databricks Runtime 20, Databricks Runtime no longer includes the bundled BigQuery JDBC driver, while Databricks Runtime 19 and earlier continue to include it. Classic compute workloads that use Spark JDBC with a jdbc:bigquery: URL fail with a driver-not-found error unless a BigQuery JDBC driver is installed.
DeepSeek V4 Pro (0813) is scheduled for retirement
DeepSeek V4 Pro (0813) will be retired on October 30, 2026. DeepSeek V4.1 Flash is the recommended replacement. See Deprecated and retired models for migration guidance.
Moonshot AI Kimi K2.7 is scheduled for retirement
Moonshot AI Kimi K2.7 will be retired on October 30, 2026. Kimi K3 is the recommended replacement. See Deprecated and retired models for migration guidance.
Base environments are now supported on classic compute (Beta)
You can now use base environments to manage Python dependencies on classic compute. You can use Databricks-provided environments or custom workspace-level environments. See Manage dependencies using environments .
Anthropic models will be removed from AWS GovCloud IL5 workspaces
On September 19, 2026, Anthropic models will be removed from Databricks on AWS GovCloud and AWS GovCloud DoD workspaces that have the DoD Impact Level 5 (IL5) compliance standard selected. After this date, IL5 workspaces can't create or use endpoints that serve Anthropic models.
Unused OAuth client secrets are automatically deleted after 90 days
Databricks now automatically deletes a service principal's OAuth client secret after 90 days without use, consistent with unused access token cleanup. If a deleted secret is still needed, create a new OAuth secret .
Automatic change data feed will soon be available in all supported regions
Automatic change data feed (Auto CDF) is rolling out to workspaces over the next month and will be available in all supported regions by the end of October 2026. Auto CDF computes row-level changes at query time using row tracking, so you don't need to enable change data feed on each Delta Lake or Apache Iceberg v3 table.