Databricks releases, every cloud
- Monitor all AI activity with the unified trace table (Beta)
Captures requests and responses in OpenTelemetry format.
- Databricks-managed MCP connectors are now integrated with Unity Gateway (Beta)
Requires reauthentication to continue using with Genie One and Genie Code.
- Environment variables for serverless jobs are in Beta
Requires workspace admin to enable from Previews page.
- Scala and Java UDFs in Unity Catalog are now generally available
Reuses existing JVM logic, offers better performance than Python UDFs.
- Protobuf tensor input for custom model serving endpoints (Public Preview)
Replaces JSON with serialized KServe v2 ModelInferRequest. Requires endpoints deployed after July 9, 2026.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud) - Object metadata column is now in Public Preview
Exposes object properties like MIME type and tags, available in Databricks Runtime 18.1.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud) - Run the change, do not just read it. Hands-on labs in your own Databricks workspace, graded when you submit. Browse labs
- MySQL integrated CDC pipeline is now in Beta
Replaces separate ingestion gateway, requires single pipeline update.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud) - Parquet v2 for Delta Lake tables is generally available
Requires Databricks Runtime 18.1 and above, set delta.parquet.format.version to 2.12.0.
- AUTO CDC flows for streaming tables in Databricks SQL are generally available
Automatically handles out-of-order records as SCD type 1 or 2, replacing complex MERGE INTO logic.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud) - Auto time-to-live for automatic row deletion is generally available
Deletes rows from Delta Lake, Iceberg, and streaming tables based on a DATE or TIMESTAMP column.
- Converting a partitioned table to liquid clustering is generally available
Uses ALTER TABLE command, requires Databricks Runtime 18.1 or above.
- Lakehouse Replay (Beta)
Automatically replays serverless workloads against upcoming runtime releases. Requires no setup.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud) - Serve custom LLMs with Custom Model Serving (Beta)
Supports multimodal models and PEFT recipes, unlike Foundation Model APIs.
- Alibaba Cloud Qwen3.5 122B A10B now available as a Databricks-hosted model in Public Preview
Accessible via Foundation Model APIs, available in Public Preview.
- Spark Declarative Pipelines on Lakeflow updates are retained for 60 days
Retention period increased from 30 to 60 days, applying to UI and REST API updates.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud) - SQL alert task for jobs (Beta)
Evaluates SQL alerts as part of workflows, enabling automated condition checks and notifications.
AWS, read it on their docsAzure, read it on their docsGCP, read it on their docsSAP(not on this cloud) - Endpoint telemetry for custom model serving endpoints (Beta)
Stores logs, traces, and metrics in Unity Catalog Delta tables using OpenTelemetry.
- AI Gateway (Beta)
Replaces Unity Gateway, adds rich UI and improved observability.
- Get this table by email. One Monday mail covering the week, filtered to the clouds and products you run. Weekly digest
- Function calling and structured output now available using the Alibaba Cloud Qwen3-Next Instruct model (Beta)
Supports Qwen3-Next 80B A3B Instruct model.
- OpenAI GPT 5 models are now generally available in Mosaic AI Model Serving
Includes GPT-5 mini and nano models.
- Alibaba Cloud Qwen3-Next Instruct now available in Beta
Supports 80B A3B Instruct in us-west-2 and ap-northeast-1 regions via pay-per-token APIs.
- Improved autoscaling behavior for Mosaic AI Model Serving
Ignores brief traffic surges, scaling on sustained load increases. Reduces unnecessary concurrency scaling.
- Mosaic AI Model Serving now supports OpenAI GPT 5 models (preview)
Supports GPT-5, GPT-5 mini, and GPT-5 nano models for batch inference with AI Functions.
- Anthropic Claude Opus 4.1 now available as a Databricks-hosted foundation model
Accessible via Foundation Model APIs on a pay-per-token basis.
- Route-optimized endpoints now require route-optimized URL path for querying
Requires route-optimized URL path for querying, standard workspace URL path not supported for new endpoints.
Headlines, dates and product areas are Databricks' own, from the release notes published for each cloud, and every item links to the note it came from. The one-line summaries are ours. Not a Databricks product and not affiliated with Databricks, Inc.