Databricks releases, every cloud
- Genie Code for AI Runtime (Public Preview)
Generates distributed training code and resolves environment issues. Requires AI Runtime.
- Upcoming end of life for the Supervisor API Coming soon
Replaced by custom agents on Databricks Apps, ends September 30, 2026.
- 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.
- Azure Databricks-provided MCP Services will GA soon Coming soon
Enables read and write operations by default for tools like Slack and GitHub, and can be disabled in Unity Gateway.
- Custom Docker images for AI Runtime CLI workloads (Beta)
Allows custom system library versions and complex dependencies, requires setup via Use custom Docker images.
- Managed agent memory is in Beta
Stores memory in Unity Catalog memory stores, accessed through REST API.
- Run the change, do not just read it. Hands-on labs in your own Databricks workspace, graded when you submit. Browse labs
- AI Runtime CLI (Beta)
Submits and manages distributed training workloads on serverless GPU compute from a local machine.
- Agent Bricks Supervisor Agent now supports web search as a built-in tool
Returns synthesized answers with source citations, also available as databricks_web_search tool in Supervisor API.
- 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.
- AI Runtime 1xH100 accelerator (Beta)
Supports 1xH100 accelerator, see Hardware options for details.
- Supervisor API for building agents (Beta)
Replaces multiple API calls with a single call to build agents.
- Agent Bricks Knowledge Assistant and Supervisor Agent are now available in additional regions
Available in ap-south-1 region, requires cross-geo processing.
- Agent Bricks now covers all Databricks AI capabilities
Covers declarative agents and all other Databricks AI capabilities.
- Classification and Information Extraction are in Public Preview
Built on v2 signatures of ai_classify and ai_extract, they provide visual UIs for document analysis.
- Supervisor Agent supports long-running tasks
Supports tasks that exceed 2-minute timeout, using task continuation.
- AI Runtime is now in Public Preview
Adds GPU support to serverless compute for deep learning workloads.
- Endpoint telemetry for custom model serving endpoints (Beta)
Stores logs, traces, and metrics in Unity Catalog Delta tables using OpenTelemetry.
- Get this table by email. One Monday mail covering the week, filtered to the clouds and products you run. Weekly digest
- Knowledge Assistant is now available by default for Enhanced Security and Compliance customers
Requires compliance security profile and HIPAA controls.
- Supervisor Agent is now available in additional regions
Now available in 9 more regions, see list of supported regions for details.
- Supervisor Agent is now Generally Available
Available in select US regions for workspaces without Enhanced Security.
- Knowledge Assistant is now available in more regions
Available in us-east-2, ca-central-1, eu-central-1, eu-west-1, ap-southeast-1, and ap-southeast-2. Requires cross-geo processing in ap-southeast-1 and ap-southeast-2.
- Knowledge Assistant is now generally available
Available in select US regions for workspaces without Enhanced Security.
- Supervisor Agent now supports Unity Catalog functions and external MCP servers
Delegates tasks to Unity Catalog functions and external MCP servers.
- 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.
- Mosaic AI Agent Framework supports automatic authentication passthrough for Lakebase resources
Requires MLFlow 3.3.2 or above.
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.