Latest releases from the Databricks ecosystem.

This release fixes an issue where tag policy and assignment operations failed for hierarchical tag keys containing slashes. It also introduces a breaking change by making the `ResourceId` field optional in `bundledeployments.Operation` and adds support for Dynamics365 connection types and `EndpointId` for vector search indexes.

The `resource_id` field in `bundledeployments.Operation` is no longer required, which is a breaking change. Several bug fixes improve performance and stability, including caching OIDC tokens, making `WorkspaceClient.dbutils` lazy, and better handling of Spark Connect runtimes.

The Databricks SDK for Go now includes a TypeOverrides field for database.SyncedTableSpec and postgres.SyncedTableSyncedTableSpec. This allows practitioners to specify type overrides when defining synced tables.

The `direct` deployment engine is now Generally Available and the default for new deployments, with existing deployments retaining their current engine. New commands `databricks quickstart` and `databricks version --check` are added, alongside fixes for authentication and bundle deployments.

1.12.1

This release exposes Databricks Jobs IDs in dbt run results and adds support for SPOG vanity URLs. It also fixes issues with streaming table diffs, column-level constraints, and managed Iceberg incremental models losing clustering.

This release adds an `AcceleratedSync` field to `database.SyncedTableSpec` and `postgres.SyncedTableSyncedTableSpec`. This new field enables configuration of accelerated sync for synced tables within Databricks.

This release introduces new services for AI Search and Bundle Deployments, along with numerous new fields across various existing services like Catalog, MLflow, and Vector Search. It also includes breaking changes by removing the `bundle` package and its associated service.

This release introduces new services for AI Search and Bundle Deployments, along with numerous new fields across existing services like Catalog, ML, and Vector Search. The `bundle` package and its associated workspace service have been removed.

This release introduces new services for AI Search and Bundle Deployments, along with several new fields across various services like Catalog, ML, and Vector Search. It also includes breaking changes by removing the old Bundle package and its associated workspace-level service.

* Canonicalize Bearer tokenType in Authorization headers

The SDK now detects the AI_AGENT environment variable for user agent reporting and passes unrecognized agent values through. Pagination has new explicit factory methods, and a bug was fixed where token-paginated responses with empty pages and a next token would silently drop results.

The SDK now better detects AI agents by honoring the Vercel AI_AGENT environment variable and passing through unrecognized agent names in the User-Agent header. This allows for more specific identification of AI agents beyond just "unknown".

This release is a backport from the 0.32.x line, which will receive voluntary support for a period. Consult the full changelog for specific user-facing changes, fixes, or breaking changes.

* direct: Fix updating the apps after the Go SDK upgrade ([#5444](https://github.com/databricks/cli/pull/5444))

Release: v2.11.0 (#1902)

This release introduces a Unity Catalog explorer and a workspace filesystem explorer, enhancing direct data and file management within VS Code. It also adds support for SPOG host URLs.

The experimental open command now supports opening various Databricks resource types via their workspace URLs. Databricks Bundles gain a new --select flag for partial deployments and improved retry logic for transient HTTP errors.

The Databricks SDK for Java now correctly handles OAuth token exchanges for browser-based flows by making the client ID optional in `DatabricksOAuthTokenSource`. This fixes a `NullPointerException` when no client ID is present, allowing account-wide token federation.

Creating an external location with file events disabled now correctly applies this setting instead of silently defaulting to enabled. Previously, the `enable_file_events = false` configuration was ignored, leading to file events being enabled by default.

Databricks Asset Bundles now support a DeploymentMode field for both deployments and versions. Workspace settings include new fields for CollaborationPlatformConnectivity and EffectiveCollaborationPlatformConnectivity.

This release adds new `deploymentMode` fields to bundle deployment and version objects. It also introduces `collaborationPlatformConnectivity` and `effectiveCollaborationPlatformConnectivity` fields to the settings API.

This release adds new `deployment_mode` fields to the bundle deployment and version services. It also introduces `collaboration_platform_connectivity` and `effective_collaboration_platform_connectivity` fields to the settings service.

MLflow 3.13.0 introduces a new Role-Based Access Control system with an Admin UI for managing users and permissions, alongside trace retention and auto-archival to object storage. This release also includes one-click observability for coding agents via the AI Gateway, new engines for MLflow Assistant, and an official Helm chart for Kubernetes deployments.

The SDK now supports a discovery flow that lands users on the account selector and includes a new method for updating workspace-level token management. Additional fields for job deployments, pipeline deployments, and token information have been added.

The SDK now uses `X-Databricks-Workspace-Id` instead of `X-Databricks-Org-Id` for workspace-scoped API calls, accepting various workspace identifier formats. Several new fields and a `updateTokenManagement()` method have been added across various services, including job and pipeline deployments, and token management.

This release adds comprehensive CRUD operations for feature engineering streams and a method to update token management settings. It also introduces breaking changes by removing `catalog_id` from `CatalogCatalogStatus` and `synced_table_id` from `SyncedTableSyncedTableStatus`.

You can now manage Git credentials for service principals and permissions for Agent Bricks resources. Key fixes include proper updates for Metastore external access, improved handling of UC object destruction, and configurable timeouts for Vector Search Index creation.

This release adds new methods for managing feature engineering streams and introduces `Parameters` fields for Databricks Jobs and Pipelines. It also includes breaking changes by removing `CatalogId` and `SyncedTableId` fields from PostgreSQL catalog status objects.

This release adds comprehensive CRUD operations for feature engineering streams and introduces a `parameters` field across various Jobs and Pipelines API objects. It also includes breaking changes by removing `catalogId` and `syncedTableId` fields from specific Postgres service objects.

Bundle users now receive a suggestion to set `bundle.engine: direct` in `databricks.yml` when encountering direct-only resource errors with the Terraform engine. The CLI also adds support for managing `vector_search_indexes` as a direct-engine bundle resource, including UC grants and destructive operation confirmations.

The config-file loader now correctly sets the profile name to "DEFAULT" when using the legacy fallback, ensuring consistent profile identification. This fixes issues where consumers deriving identifiers from the profile name, like the Databricks CLI's OAuth cache, could not match login and read flows.

This release switches the SDK's internal formatter and linter to Ruff, aligning with Databricks' internal Python formatting guidelines. This change has no behavioral impact on the published SDK for users.

The SDK now supports Vercel's AI_AGENT environment variable for user-agent detection and passes unrecognized agent names as-is. New API fields were added for Lakeview dashboards, apps, ML materialized features, and Postgres synced tables.

This release introduces new methods for Lakeview and Postgres services, including `revert()` for Lakeview and `undeleteBranch()` for Postgres, alongside new fields for Jobs, IAM, and ML services. Several breaking changes require `actionType` and `resourceId` for bundle operations, `cliVersion` for bundle versions, and alter the `tags` field for marketplace listings and pagination for cluster events.

This release introduces a new bundle service and adds `revert()` for Lakeview dashboards and `undelete_branch()` for Postgres. It also includes breaking changes to the `tags` field in Marketplace listings and pagination for Cluster events.

MLflow now features a major overhaul of Role-Based Access Control with a new Admin UI and unified permission APIs, alongside new capabilities for GenAI agent observability including coding-agent tracing plugins and automated stress-testing. Databricks practitioners can also leverage new trace archival, Helm charts for Kubernetes deployment, OpenTelemetry span links, and database replica routing for scaling.

This release introduces new fields for IAM user attributes, job trigger state, and enhanced branch management capabilities including undelete and purge options. Several breaking changes require `ActionType`, `ResourceId`, and `CliVersion` fields in bundle operations and versions, and modify marketplace listing tags and cluster event pagination.

OAuth tokens for interactive logins are now stored in OS-native secure stores by default, requiring re-authentication after upgrade. A new `databricks aitools` command group is added for installing Databricks skills into coding agents.

Release: v2.10.8 (#1899)

- Bump Databricks JS SDK to 0.17.0

This release introduces a new bundle package and workspace-level service for managing Databricks Asset Bundles. It also adds an MtlsConfig field to ml.AuthConfig for mTLS authentication configurations.

* Add `com.databricks.sdk.service.bundle` package.

This release adds support for the VARIANT type. No other user-facing features, fixes, or breaking changes are included.

This release fixes several regressions, including issues with MERGE operations, schema overwrites with predicates, and partition column changes. It also enables passing non-string datatypes in custom commit metadata and updates the minimum PyArrow version to 21.0.0 for preliminary variant type support.

Databricks SDK for Go now supports new fields for configuring pipeline refresh selections within jobs. Additionally, new fields for operational email custom recipients are available in settings.

The Databricks SDK for Java now includes new fields for pipeline refresh selection within job parameters and tasks. Additionally, new fields for operational email custom recipients are available in the settings service.

This release adds new fields for pipeline refresh and checkpoint selection within Databricks Jobs, enhancing control over DLT pipeline execution. It also introduces new fields for custom operational email recipients in Databricks settings.

* Add `CreateWorkspaceAssignmentDetail`, `DeleteWorkspaceAssignmentDetail`, `GetWorkspaceAssignmentDetail`, `ListWorkspaceAssignmentDetails` and `UpdateWorkspaceAssignmentDetail` methods for [a.AccountIamV2](https://pkg.go.dev/github.com/databricks/databricks-sdk-go/service/iamv2#AccountIamV2API) account-level service.

* Add `createWorkspaceAssignmentDetail()`, `deleteWorkspaceAssignmentDetail()`, `getWorkspaceAssignmentDetail()`, `listWorkspaceAssignmentDetails()` and `updateWorkspaceAssignmentDetail()` methods for `accountClient.accountIamV2()` service.

This release adds new methods for managing IAMv2 workspace assignments at both account and workspace levels, and introduces new fields for ML features, disaster recovery stable URLs, and customer-facing ingress network policies. It includes breaking changes to the `list_features()` method and `ListFeaturesRequest` by adding new required catalog and schema name fields.

- Add support for metric views as a materialization by @benc-db in [#1285](https://github.com/databricks/dbt-databricks/pull/1285)

This release fixes a bug preventing partition column changes when overwriting tables and addresses a regression in MERGE operations that caused high memory usage. It also adds support for passing non-string datatypes in custom commit metadata and includes nanosecond timestamp support.

TypeScript SDK 0.2.0

Bump several RC TypeScript packages stable version.

The `vector_search_endpoints` configuration and commands now use `target_qps` instead of `min_qps`, requiring updates to `databricks.yml` and CLI invocations. Authentication commands received several usability improvements, including better keyring handling and interactive profile selection.

* Add `catalogName`, `createdAt`, `createdBy`, `name` and `schemaName` fields for `com.databricks.sdk.service.ml.Feature`.

Job tasks can now be marked as disabled in both run and submit requests. Catalog connection types gained new enum values for HubSpot, GitHub, Outlook, and Smartsheet, while the `unspecified_resource_name` field was removed from `RequestedResource` in the Postgres service.

This release fixes state decoding errors for databricks_library, databricks_share, and databricks_quality_monitor after upgrading from v1.113.0 to v1.114.0. It also resolves issues where several account-level data sources and settings failed due to workspace ID resolution errors, now supporting account-level usage correctly.

This release fixes an issue where the Databricks CLI `--profile` fallback was broken. It also introduces new API methods for `workspaceClient.supervisorAgents()` and `workspaceClient.vectorSearchEndpoints()`, along with several breaking changes related to `Example` and `Tool` fields, and `minQps` in Vector Search endpoints.

This release introduces new methods for managing supervisor agents and vector search endpoints, along with several new fields for connector options, ingestion sources, and customer-managed keys. Breaking changes include the removal of `min_qps` fields from vector search endpoint configurations and making `guidelines` and `description` fields optional in certain services.

The default query comment now includes the invocation_id. This release also fixes several issues, including enforcing the 255-character identifier limit for Databricks relations and suppressing various spurious warnings.

* Add `ConfluenceOptions` field for [pipelines.ConnectorOptions](https://pkg.go.dev/github.com/databricks/databricks-sdk-go/service/pipelines#ConnectorOptions).

Release: v2.10.7 (#1895)

This release adds initial compatibility for Remote Development and renames "Databricks Asset Bundles" to "Declarative Automation Bundles." It also includes fixes for profile management, such as preserving profile names during CLI authentication and correctly listing profiles with account IDs.

The `databricks api` command now supports unified hosts, allowing `--account` and `--workspace-id` flags or `?o=<workspace-id>` URL parameters for routing. JSON output for single objects now uses standard key-value spacing.

Nightly - unity-catalog

Nightly build from unity-catalog

MLflow 3.12.0 introduces multimodal tracing, allowing users to store and render PDFs, audio, and images as artifact attachments in tracing spans, along with new support for Codex, Gemini, and Qwen coding agent tracing. Additionally, it enables setting guardrails on gateway endpoints to prevent unsafe model inputs/outputs and deprecates `enable_mlserver` in the pyfunc serving backend.

Nightly - fix-new-profile-sign-in

Nightly build from fix-new-profile-sign-in

You can now adopt pre-existing Databricks Postgres branch and endpoint resources using the replace_existing argument. A change to databricks_external_location prevents Terraform drift related to the effective_file_event_queue field.

This release introduces new services for disaster recovery and temporary volume credentials, alongside extensive new methods for knowledge assistants. It also adds support for various new fields and enum values across several services, while making the `description` field optional for `SupervisorAgent` and removing the `connection` field from `Tool`.

* Fixed Databricks CLI `--profile` fallback by detecting the CLI version at init time. The previous error-based detection was broken because `--profile` is a global Cobra flag silently accepted by old CLIs.

The `--experimental-is-unified-host` flag and related configuration options are removed; unified hosts are now detected automatically. List commands now support interactive pagination, displaying 50 rows at a time with options to view more or quit. An experimental opt-in for OS-native secure token storage is introduced via `DATABRICKS_AUTH_STORAGE=secure`. A panic in `databricks warehouses update-default-warehouse-override` when missing arguments is fixed, and relative paths in `alert_task.workspace_path` for job tasks are now translated to fully qualified workspace paths.

Rolls back to v1.113.0 as the stable release.

This release introduces new resources and data sources for managing disaster recovery failover groups, stable URLs, supervisor agents, and UC secrets. It also enables adopting pre-existing Postgres branch and endpoint resources using the `replace_existing = true` argument.

MLflow 3.12.0rc0 introduces enhanced AI agent development features, including automatic tracing for various AI coding assistants and OpenClaw, along with new AI Gateway Guardrails for safety checks. It also adds multimodal trace attachments for viewing images, audio, and files directly in the MLflow UI, and a new `mlflow.diffusers` flavor for saving and serving diffusion models.

This release introduces a new disaster recovery package and adds methods for managing knowledge assistant examples. It also includes breaking changes related to the `supervisoragents.Connection` and `supervisoragents.Tool` fields.

Unity Catalog AI 0.4.0

DatabricksFunctionClient now supports an optional warehouse_id for function execution, enabling use in workspaces without serverless compute. Python 3.10+ is now required, and several bug fixes address issues with Gemini toolkit, LangGraph, and OSS client function creation.

Unified host detection is now automatic, removing the `Experimental_IsUnifiedHost` field and enabling a single configuration profile for both account and workspace operations. The file-based OAuth token cache has been removed, defaulting to an in-memory cache unless a persistent cache is explicitly provided.

The SDK now automatically detects AI coding agents and appends agent information to HTTP request headers, while also removing the unused `experimentalIsUnifiedHost` field from `DatabricksConfig`. A bug fix addresses `X-Databricks-Org-Id` header issues for `SharesExtImpl.list()` on SPOG hosts, and several API method paths have changed, which are breaking changes.

This release introduces new workspace-level services for secrets_uc and supervisor_agents, along with an update method for tokens. Several breaking changes occurred, primarily due to method path changes for data classification, environments, knowledge assistants, postgres, and warehouses services.

WorkspaceExt upload/download and SharesExt list now include the X-Databricks-Org-Id header for SPOG host compatibility. WorkspaceClient.get_workspace_id avoids an API call when the workspace ID is already known, fixing a SPOG host failure.

The CLI now supports a --limit flag for paginated list commands and caches host metadata lookups for faster repeated invocations. Bundles gain support for Vector Search Endpoints and prompt for confirmation before destroying Lakebase resources.

This release enhances the new Datafusion TableProvider and improves log parsing performance. It also fixes numerous bugs, including issues with DeltaScan schema handling, streamed merges, and incorrect row counts for DELETE operations.

This release introduces new workspace-level services for supervisor agents and Unity Catalog secrets, along with an update method for tokens. Several existing API methods for data classification, environments, knowledge assistants, Postgres, and warehouses have breaking changes due to path modifications.

This release adds Azure MSI authentication support and improves `.databrickscfg` default profile resolution. It also fixes issues with non-JSON error responses and Databricks CLI token scope mismatches, alongside several API additions and two breaking changes.

The SDK now supports a default host metadata resolver for easier shared configuration and improved SPOG host compatibility for Workspace and Shares API calls. WorkspaceClient.CurrentWorkspaceID() directly uses the configured WorkspaceID, eliminating an API call and fixing SPOG host authentication issues.

Nightly - fix/replace-softprops-action

Nightly build from fix/replace-softprops-action

This release automatically detects unified hosts for account and workspace operations, and accepts DATABRICKS_OIDC_TOKEN_FILEPATH for OIDC token files. Python 3.8 and 3.9 are no longer supported, requiring Python 3.10 or newer.

This release adds support for Notebook-scoped packages when submitting commands or running notebook jobs. It also includes fixes for workflow job creation, preventing duplicate aliases in empty mode, and enabling insert-by-name for microbatch and replace_where strategies.

This release fixes an error where bundle commands failed due to an expired key when downloading Terraform. Users will no longer encounter the "unable to verify checksums signature: openpgp: key expired" error.

This release fixes an error that prevented bundle commands from running due to an expired key when downloading Terraform. Users can now execute bundle commands without encountering checksum signature verification failures.

This release fixes an error where bundle commands failed due to an expired key when downloading Terraform. Users can now run bundle commands without encountering checksum signature verification issues.

This release fixes an error where bundle commands failed due to an expired OpenPGP key during Terraform downloads. Users can now run bundle commands without encountering checksum verification errors.

This release fixes an error where bundle commands failed due to an expired key when downloading Terraform. Users will no longer encounter "unable to verify checksums signature: openpgp: key expired" errors.

This release fixes a critical error preventing bundle commands from downloading Terraform due to an expired key signature. Users previously encountering "unable to verify checksums signature: openpgp: key expired" will now find bundle commands functional.

This release fixes an error where bundle commands failed due to an expired key when downloading Terraform. Users can now run bundle commands without encountering checksum signature verification issues.

This release fixes a critical error preventing bundle commands from running due to an expired GPG key. Databricks practitioners can now use bundle commands without encountering checksum verification failures.

Databricks CLI v0.297.2 fixes a critical "key expired" error that prevented `databricks bundle deploy` from downloading Terraform binaries. This was resolved by hardcoding the ArmoredPublicKey used for Terraform binary installation.

UnityCatalog 0.4.1

The Unity Catalog Spark connector now supports atomic REPLACE TABLE AS SELECT and Dynamic Partition Overwrite for managed Delta tables, and a credential-scoped file system to prevent out-of-memory errors. This release also adds support for the VARIANT data type in the UC client and fixes a critical security vulnerability (CVE-2026-27478) related to JWT issuer validation, which is a breaking change requiring new server configurations.

This release adds new resources for managing Postgres catalogs, synced tables, and workspace base environments. It also introduces an `api` field for dual account/workspace resources to explicitly control API level, supporting unified hosts.

Delta Lake 4.2.0

This release enhances Unity Catalog managed tables with support for REPLACE TABLE, RTAS, Dynamic Partition Overwrite, and improved streaming read options like `startingTimestamp`. It also introduces experimental Kernel-based Flink connector for catalog-managed tables and generally available Variant and Geospatial types for Delta Kernel.

This release raises the minimum Go version to 1.24 and introduces new features like a host metadata resolver hook and a lazy iteration utility with item limits. It also includes numerous bug fixes for token acquisition and caching, along with extensive API additions for various Databricks services.

TypeScript SDK 0.2.0 RC1

Release candidate for `@mlflow/vercel` TypeScript package with version 0.2.0: https://github.com/mlflow/mlflow/pull/22105

MLflow 3.11.1 introduces AI-powered issue detection for agent traces, budget alerts and limits for AI Gateway spending, and a new interactive graph view for visualizing trace hierarchies. It also enhances security with pickle-free model serialization and improves dependency management with native UV support.

Model Catalog

Per-provider model catalog files. Updated weekly by CI.

Stripped third-party dependencies from evaluation and AI Gateway features, replacing external provider routing with built-in implementations.

This release adds a new method to force-refresh cached U2M OAuth tokens, returning an error on failure instead of falling back. It also introduces a new error type for when a refresh token is missing during a refresh request.

This release adds a new `disableGovTagCreation` field to the `RestrictWorkspaceAdminsMessage` for both settings and settingsv2 services. This field allows configuring restrictions on the creation of governance tags by workspace administrators.

The SDK now automatically detects AI coding agents and appends `agent/<name>` to HTTP request headers. Two new `DisableGovTagCreation` fields were added to `RestrictWorkspaceAdminsMessage` in both `settings` and `settingsv2`.

The SDK now automatically detects AI coding agents and appends agent information to HTTP request headers. Two new `disable_gov_tag_creation` fields were added to restrict workspace admin settings.

* Add resource and data sources for `databricks_postgres_role`.

* Add `parentPath` field for `com.databricks.sdk.service.dashboards.GenieSpace`.

OAuth token refreshing is now proactive for tokens expiring within five minutes, improving token validity for callers. The `dashboards.GenieSpace` struct gains a new `ParentPath` field.

This release introduces a new `environments` service for managing Databricks environments and adds a `parent_path` field to `GenieSpace` dashboards. It also includes a new `can_create_app` permission level for IAM.

The Databricks SDK for Go now supports specifying a `default_profile` within the `[__settings__]` section of your `.databrickscfg` file. This allows for easier configuration management when working with multiple profiles.

MLflow 3.11.0rc0 introduces AI-powered issue identification for agent traces, budget alerts for AI Gateway spending, and a new interactive graph view for trace hierarchies. It also adds pickle-free model serialization for enhanced security and native OpenTelemetry GenAI convention support for trace exports.

Release: v2.10.6 (#1858)

This release fixes an issue where the VS Code extension would not correctly handle 404 errors from the Databricks SDK. This improves stability when interacting with Databricks resources.

Databricks Jobs now support new alert-related fields in run outputs and task definitions, enabling more granular notification configurations. A new `environments` package and service were introduced, along with a `CAN_CREATE_APP` permission level for IAM.

This release adds new fields for job alert configurations across various job-related structs, including `RunOutput`, `RunTask`, `SubmitTask`, and `Task`. It also introduces a new `environments` package and service for managing workspace environments, alongside a `CanCreateApp` permission level for IAM.

You can now specify a `default_profile` in your `.databrickscfg` file's `[__settings__]` section for consistent profile resolution. New `alert_output` and `alert_task` fields have been added to several job-related API objects.

python-v1.5.0: faster writes, log compaction, spil config in MERGE

This release introduces faster writes through parallel partition writers, log compaction, and disk spilling for MERGE operations. A breaking change alters get_add_actions to return an ArrowTable instead of an ArrowRecordBatch.

This release adds new fields to the Ingestion Pipeline Definition and Origin objects, enhancing configuration options for data ingestion pipelines. It also introduces a subDomain field for External Function Requests in the serving API.

This release adds new fields and methods across several services, including updates for ML features, pipelines, and Postgres roles. It also introduces breaking changes by making previously required fields optional in ML-related DeltaTableSource, Feature, Function, and KafkaSource configurations.

Databricks SDK for Python now supports new fields for defining ingestion pipelines, including connector type, data staging options, and detailed ingestion source information. External function requests in model serving can now specify a sub-domain.

This release enables concurrent microbatch execution and adds an optimize() call to snapshot materialization. It also fixes issues with quoted catalog names, streaming table alter SQL, missing optimize() calls for table v2, column-level tags for V1 tables, and constraint enforcement.

This release adds new fields and methods for managing ML features, functions, and Kafka sources, including a new updateRole method for workspace Postgres services. Several fields related to ML DeltaTableSource, Feature, Function, and KafkaSource are now optional, which is a breaking change.

This release introduces new methods for the Genie workspace service and an update_role method for the Postgres service. Several fields across ML and other services are now optional, including breaking changes for `entity_columns`, `inputs`, `function_type`, `entity_column_identifiers`, and `timeseries_column_identifier` in various ML-related services.

This release introduces new resources for managing Postgres databases, data classification catalog configurations, and knowledge assistant features. It also renames the `databricks_apps_space` resource to `databricks_app_space`.

This release introduces new services for Data Classification and Knowledge Assistants, accessible via `workspaceClient`. It also adds several new methods for managing Genie evaluation runs and results.

Databricks SDK for Java now allows fine-grained control over HTTP request timeouts through a new `withRequestConfig` method on `CommonsHttpClient.Builder`. This enables practitioners to configure specific timeout settings for their API calls.

Databricks CLI authentication now correctly errors on token scope mismatches, prompting re-authentication instead of silently using incorrect permissions. New `dataclassification` and `knowledgeassistants` services and corresponding workspace-level APIs have been added.

This release adds a "try-it" page for Gateway usage examples and filters gateway experiments from the experiment list in the UI. It also fixes numerous UI issues, artifact download problems, and tracing errors, including issues with model copying across workspaces.

The SDK now dynamically adjusts authentication token stale periods, extending them up to 20 minutes for standard OAuth. Users can override this behavior by manually setting the stale period in the CachedTokenSource builder.

The SDK now dynamically adjusts authentication token refresh periods based on the token's initial lifetime, up to 20 minutes for standard OAuth. Users can override this behavior by specifying a fixed stale_duration in the Refreshable class constructor.

This release introduces row filter functionality and support for metric views.

Delta Lake 4.1.0

Delta Lake 4.1.0 introduces enhanced support for Unity Catalog managed tables, including atomic CTAS and conflict-free feature enablement for Deletion Vectors and Column Mapping. It also drops support for Spark 3.5, now requiring Spark 4.0.1 or higher, and mandates Java 17.

Utility clusters created by resources like `databricks_aws_s3_mount` now default to `SPOT_WITH_FALLBACK` for improved reliability. Plaintext credential fields in `databricks_model_serving` and `databricks_git_credential` are now marked sensitive to prevent display in plan/apply output.

MLflow 3.10.0 introduces multi-workspace support for organizing experiments and models, alongside new GenAI features like multi-turn evaluation, LLM cost tracking, and AI Gateway usage analytics. The UI has been redesigned for improved navigation, and in-UI trace evaluation is now available.

* Mark `effective_enable_file_events` as read-only in `databricks_external_location` to prevent Terraform drift.

This release fixes an issue where multiple foreign keys between tables were not retained after an incremental run. It also resolves a bug where changes to materialized view partition_by clauses failed, now using a DROP and CREATE strategy.

You can now manage Lakebase database project permissions using `database_project_name` in `databricks_permissions` and configure instance pool node type flexibility with a new block in `databricks_instance_pool`. A bug was fixed that previously caused errors during WorkspaceClient() creation in `databricks_grant` and `databricks_grants` resources.

This release adds new resources and data sources for managing Databricks Apps Space and Endpoints. It also updates the underlying Go SDK to version 0.108.0.

This release provides a fix for a Rust-related issue, backporting a previous pull request to the Python v1.3.3 version. Databricks users will experience improved stability or corrected behavior due to this specific fix.

UnityCatalog 0.4.0

Unity Catalog now fully supports AWS Storage Credentials and External Locations, enabling secure, governed access to S3 data via temporary, scoped IAM roles. Credential renewal for cloud storage is now enabled by default in the Spark connector, which also gains atomic CTAS for Delta tables and support for Spark 4.1/Delta 4.1.

The `databricks_workspace_file` resource now supports payloads larger than 10MB, and `databricks_mws_storage_configurations` includes a `role_arn` field for S3 bucket sharing with Unity Catalog. Several bug fixes address issues with `databricks_mws_ncc_private_endpoint_rule` updates, `databricks_secret_acl` management, `databricks_app` resource reading, and `databricks_users` data source `extra_attributes` parameter.

MLflow now supports multi-workspace environments for organizing experiments and resources, alongside a new top-level navigation split for GenAI and Classical ML workflows. Key new features include multi-turn conversation simulation, automatic LLM trace cost tracking, AI Gateway usage analytics, and a CLI command to generate a demo environment.

This release introduces a session-first DataFusion integration and exposes Delta Lake Vacuum (DV) metadata as Arrow streams. It also fixes issues with schema merge appends for generated columns and improves parquet predicate pushdown.

This release fixes issues where file statistics did not align with the Parquet read schema and DML predicates were not resolved against the execution scan schema. It also updates the asserted nullability in the DataValidation output schema.

SQL warehouses now support "5X-Large" cluster sizes and a higher maximum of 40 clusters. This release also fixes permanent drift for external model credentials in databricks_model_serving and improves dashboard file content change detection.

Release: v2.10.5 (#1834)

- Update Databricks CLI to v0.286.0

v.3.9.0

MLflow 3.9.0 introduces an in-product MLflow Assistant chatbot and a Trace Overview Dashboard for GenAI experiments, enhancing debugging and performance insights. The AI Gateway is revamped for direct tracking server integration, alongside new LLM judge features for online monitoring and custom prompt building.

This release enables deletion vector features for Delta tables and improves logical planning for update and delete operations. It also fixes issues with reporting failed data in data checks and supporting user names in Azure URLs.

This release adds new resources for account user settings, default warehouse overrides, and fixes issues with importing databricks_share and creating databricks_dashboard resources. The exporter now supports additional network policy resources and rewrites cloud-specific attributes in cluster policies.

This release updates the dbt-core pin for the 1.10.latest version. No user-facing features, fixes, or breaking changes are included.