In the last seven days: 12 shipped, seven in preview, five with a date on it and six new concepts on Lakenaut.

24 changes for you: five needs action, eight worth evaluating and 11 nice to know.

New features

• Serverless compute access controls reach GA; rate limits and compute tags in Beta Databricks says Serverless Compute Access Controls are generally available: admins grant Can Use or Can Manage on the default interactive and automated serverless compute objects.

• Real-Time Mode reaches GA for Lakeflow pipelines, with ForEach sinks to operational stores Real-Time Mode on Spark Declarative Pipelines (Lakeflow pipelines) is generally available, giving millisecond-range latency for streaming flows.

• Dashboard relationships for multi-fact data models reach general availability They let an author build multi-fact data models scoped to a single AI/BI dashboard.

• Clean room notebooks can run on serverless GPU, now generally available Clean room notebooks can now run on serverless GPU compute for model inference and training over shared data, on AI Runtime, with the same approval flow.

• Unity Catalog permissions on system.ai now control which AI Functions users can run (GA) Metastore admins can use Unity Catalog permissions on the system.ai schema to decide which task-specific AI Functions people in the organization may run.

• Lakeflow Jobs show job-level query performance metrics and insights (GA) Improved Lakeflow Performance Observability is generally available.

• AI Search can index Streaming Tables, Materialized Views and managed Iceberg tables directly You create a Delta Sync index pointing at the source table, so the intermediate Delta copy and custom synchronization pipeline are no longer needed.

• Unity Catalog volume files can be edited in the workspace file editor (GA) Files stored in Unity Catalog volumes can now be opened and edited in place in the workspace file editor, with no download and re-upload.

• Default OAuth client IDs let ChatGPT, Claude, Cursor and others reach Databricks MCP servers An account admin no longer has to create an OAuth app first.

• Unity Gateway Usage Dashboard v1.0 covers coding-agent adoption and spend It shows coding-agent adoption, harness usage and estimated session spend, with wider model and user spend analysis and spend attribution by tag.

• New account-level security features are available to all accounts as soon as they reach GA Account-level security features, such as private connectivity, ingress and egress controls and encryption.

• Spend Status API reports usage against Unity Gateway budget limits It returns the status of a single budget or lists statuses across budgets.

Previews and beta

• DataFrame caching on serverless compute enters Public Preview df.cache() and df.persist() now work on serverless compute, so an expensive result can be reused for the rest of a session.

• Lakeflow Connect adds a Microsoft Graph Logs connector in Beta for security alerts It ingests security alerts from Microsoft Defender XDR, Microsoft Sentinel and other Microsoft security products.

• Lakeflow Connect adds a Xero connector in Beta for accounting data It ingests accounting data such as invoices, payments and contacts, using OAuth user authorization.

• Several real-time flows can share one serverless pipeline; both new options are in Beta On serverless pipelines using the preview channel, multiple real-time mode flows can run in one pipeline without competing for task slots.

• Lakeflow Connect adds a managed QuickBooks connector in Beta It ingests accounting data from QuickBooks Online, such as accounts, customers, vendors, invoices, bills, payments and journal entries.

• Lakeflow Connect adds a managed PayPal connector in Beta It ingests payments data such as transactions, disputes, invoices, balances, catalog products and billing plans.

• Lakeflow Connect adds a Box Audit Logs connector in Beta It ingests enterprise audit events from Box into Databricks.

Deprecations and end of support

• DeepSeek V4 Flash (0731) will be retired on 5 November 2026 DeepSeek V4.1 Flash is the recommended replacement.

• Databricks Runtime 10.4 LTS ends on 2026-11-01 Anything still relying on it needs a plan before then.

• Genie One MCP Beta endpoint (/api/2.0/mcp/genie) ends on 2026-10-31 Anything still relying on it needs a plan before then.

• DeepSeek V4 Pro (0813) ends on 2026-10-30 Anything still relying on it needs a plan before then.

• Genie One MCP Beta endpoint now ends 30 November 2026; move to Unity Gateway Workloads that call it need to switch to the generally available system.ai.genie_one_mcp MCP in Unity Gateway. Action required, by 30 Nov 2026: Point every client of the Beta Genie One MCP endpoint at the system.ai.genie_one_mcp MCP in Unity Gateway.

New and rewritten concepts

• Access modes: standard and dedicated Standard access mode shares one compute resource between isolated users; dedicated assigns it to one user or group. What each allows, which Unity Catalog features need which.

• Choosing compute: all-purpose, job cluster, serverless, SQL warehouse Databricks offers serverless compute, all-purpose clusters, job clusters, and SQL warehouses. Each has its own DBU-based cost model, its own limits.

• The CLI and the SDKs The Databricks CLI, the language SDKs, and Databricks Connect all share one authentication order and wrap the same REST API for scripting and automation.

• Coding agents on Databricks, and how to keep them safe Agent skills teach a coding agent like Claude Code the current way to work on Databricks; the identity you give it, not its instructions, decides what it can break.

• AI Runtime and serverless GPU compute Serverless GPU compute for training and fine-tuning: a @distributed decorator in notebooks, the air CLI with a workload YAML, and the successor to Foundation Model Fine-tuning.

• AI Search index types and sync modes The four AI Search index options, continuous against triggered sync, and standard against storage-optimized endpoints, with the choices you cannot undo later.