settings-v2.md

Databricks settings v2 (/api/2.1/settings/{name})

Auto-generated on 2026-07-06T07:33:57Z. Catalog from GET /api/2.1/settings-metadata; status column from per-name probe against a Databricks host. 153 entries. Use with the databricks_workspace_setting_v2 Terraform resource.

Status legend: ✅ 200 set · 🟡 404 recognized but unset · ❌ other failure · - not probed

Index by preview phase

• BETA: 66 settings
• GA: 49 settings
• GA_SOON: 4 settings
• PRIVATE_PREVIEW: 1 settings
• PUBLIC_PREVIEW: 33 settings

Summary

Name	Phase	Status	Display Name
agent_monitoring	BETA	✅ 200	Production Monitoring for MLflow
agents_obo	PUBLIC_PREVIEW	✅ 200	Agent Framework: On-Behalf-Of-User Authorization
aha_connector	BETA	🟡 404	Lakeflow Connect for Aha!
ai_classify	GA	✅ 200	AI Classify
ai_extract	GA	✅ 200	AI Extract
ai_parse_document	GA	✅ 200	AI ParseDocument
ai_prep_search	BETA	🟡 404	AI Prep Search
ai_runtime_beta_features	BETA	🟡 404	AI Runtime Beta Features
air_h100_multinode	BETA	✅ 200	Serverless GPU Compute API Remote H100s
air_interactive	PUBLIC_PREVIEW	✅ 200	Serverless GPU Compute
alerts_v2	GA	✅ 200	SQL Alerts V2
alertv2_job_task	GA	✅ 200	Alert Job Task
allowedAppsUserApiScopes	GA	✅ 200	-
anomaly_detection_ws	PUBLIC_PREVIEW	✅ 200	Data quality monitoring with anomaly detection (workspace level)
apps_otel	PUBLIC_PREVIEW	✅ 200	OpenTelemetry for Databricks Apps
apps_v2_ui	GA	🟡 404	Databricks Apps V2
authoring_context	PUBLIC_PREVIEW	✅ 200	Focused notebook & file editor for Git folders
cld_to_volumes	GA	✅ 200	Cluster Log Delivery to UC Volumes
cloudfiles_excel	GA	✅ 200	Excel File Format Support
collaboration_platform_connectivity	GA	✅ 200	-
collaboration_platform_message_visibility	GA	✅ 200	-
confluence_connector	GA	✅ 200	Lakeflow Connect for Confluence
conn_cdc_col_select	PUBLIC_PREVIEW	✅ 200	Lakeflow Connect Column Selection for Database Sources
custom_apps_preview	GA	✅ 200	Databricks Apps
custom_llm_serving	BETA	✅ 200	Custom LLM Serving for Databricks Model Serving
customerApprovedWSLoginExpirationTime	GA	✅ 200	-
dabs_templates	BETA	✅ 200	Custom bundle templates in the workspace
dabs_visual_edit	BETA	✅ 200	Visual authoring: UI <> YAML Sync for DABs in the Workspace
dashboard_authoring_agent	GA	🟡 404	Genie Code for dashboard authoring
databricks_one	GA	✅ 200	Databricks One
dataclassificationws	GA	✅ 200	Data Classification
datascience_agent	GA	✅ 200	Genie Code
dbsql_5xl_pro	PUBLIC_PREVIEW	✅ 200	5XL PRO SQL Warehouse
dbsql_5xl_serverless	PUBLIC_PREVIEW	✅ 200	5XL Serverless SQL Warehouse
dbt_cloud_task	BETA	✅ 200	dbt platform task
dc_custom_tags	BETA	🟡 404	Extend Data Classification with Custom Classifiers
default_base_envs	GA	✅ 200	Workspace base environments for serverless compute
default_wh_setting	GA	✅ 200	Default warehouse setting
designer	GA	✅ 200	Lakeflow Designer
direct_cdc_connector	BETA	✅ 200	LakeFlow Connect for Direct Cdc Managed Ingestion Pipeline
disable_legacy_access	GA	✅ 200	-
disable_legacy_dbfs	GA	✅ 200	-
discover_page	PUBLIC_PREVIEW	✅ 200	Discover Page
ds_v2_join_pushdown	PUBLIC_PREVIEW	✅ 200	Join Pushdown for Federated Queries
dynamics_connector	PUBLIC_PREVIEW	✅ 200	Lakeflow Connect for Dynamics 365
embedded_genie	GA	🟡 404	Embed Genie as an iframe
enable_dcs_vnext	BETA	✅ 200	DCS-vNext
enable_dlmv_aibi	PUBLIC_PREVIEW	✅ 200	Enable Dashboard Local Metric Views in AI/BI Dashboards
enable_github_webhook_app_deployments	BETA	🟡 404	Databricks Apps - GitHub Webhook-Triggered Deployments
enable_lakeview_tags	PUBLIC_PREVIEW	✅ 200	Tagging support for workspace scoped assets
enable_obo_user_apps	GA_SOON	✅ 200	Databricks Apps - On-Behalf-Of User Authorization
enforceGitAppDeployments	GA	✅ 200	-
excel_connector_flip	PUBLIC_PREVIEW	🟡 404	Excel Add-In
exp_sp_token_notif	BETA	✅ 200	Expiring service principal access token notifications
external_access_to_managed_delta	PUBLIC_PREVIEW	🟡 404	External Access to Unity Catalog Managed Delta Table
external_engine_fgac	BETA	🟡 404	Cross-engine ABAC
filebrowser_tree	PUBLIC_PREVIEW	✅ 200	Tree view of the side panel file browser
fmapi_qwen3_instruct	PUBLIC_PREVIEW	✅ 200	Enable Extended Models
foreign_tbl_comments	BETA	✅ 200	Comments on Foreign Tables
fstore_decl_fw	PUBLIC_PREVIEW	✅ 200	Feature Views (Batch)
fstore_decl_strm_fw	PUBLIC_PREVIEW	✅ 200	Feature Store Streaming Feature Views
full_screen_genie_code	BETA	🟡 404	Full Page Genie Code
full_text_search_index	BETA	🟡 404	SQL: Full-Text Search Index for UC managed tables
gdrive_connector	BETA	✅ 200	Lakeflow Connect for Google Drive
generic_lfc	BETA	✅ 200	Lakeflow Connect Community Connectors
genie_bi_migration	BETA	🟡 404	Import from External BI to AI/BI
genie_chat_sharing	BETA	✅ 200	Genie Chat Sharing
genie_deep_research	GA	✅ 200	Genie Agent
genie_inspect_answer	PUBLIC_PREVIEW	✅ 200	Genie Answer Inspection
genie_unstructured_files_in_deepresearch_mode	BETA	🟡 404	Upload Local PDFs to Genie Spaces
github_connector	BETA	✅ 200	Lakeflow Connect for Github
google_ads_connector	BETA	✅ 200	Lakeflow Connect for Google Ads
hspot_mktg_connector	GA	✅ 200	Lakeflow Connect for HubSpot
icebergv3	GA	✅ 200	Iceberg V3
ip_functions	PUBLIC_PREVIEW	🟡 404	Ip Functions
jdbc_connector	PUBLIC_PREVIEW	✅ 200	Custom JDBC on UC Compute
jdbc_oauth_m2m_connector	BETA	🟡 404	OAuth M2M Support for Custom JDBC on UC Compute
jira_connector	BETA	✅ 200	Lakeflow Connect for Jira
jobs_disabled_tasks	GA	✅ 200	Disabled tasks in Lakeflow Jobs
jobs_serverless_managed_base_environments	BETA	🟡 404	Serverless workspace base environment support in Jobs
lakebase_accel_sync	PRIVATE_PREVIEW	✅ 200	Lakebase Accelerated Sync
lakebase_cdf	PUBLIC_PREVIEW	✅ 200	Lakebase CDF
lakebase_otel_integration	BETA	🟡 404	Lakebase OpenTelemetry Integration
lakebase_search	BETA	🟡 404	Lakebase Search
lakeflow_new_jobs_ui	GA	✅ 200	Lakeflow Jobs UI
lakeflow_qbc	GA	✅ 200	Lakeflow Connect Query Based Connectors
lakeflow_runs_list	GA	✅ 200	Unified Runs List
lakehouse_replay	BETA	✅ 200	Lakehouse Replay
lf_pipelines_auth	GA	✅ 200	Lakeflow Pipelines Editor
llm_proxy_partner_powered	GA	✅ 200	-
managed_mcp_servers	PUBLIC_PREVIEW	✅ 200	Managed MCP Servers
managed_memory_agents	BETA	🟡 404	Managed Memory for Agents
marketplace_app_install	PUBLIC_PREVIEW	🟡 404	Marketplace - Install Databricks Apps
meta_ads_connector	BETA	✅ 200	Lakeflow Connect for Meta Ads
mlflow_logged_models	PUBLIC_PREVIEW	✅ 200	Models in Unity Catalog: Deployment Jobs
model_triggers	BETA	✅ 200	Model update job triggers
monday_com_connector	BETA	🟡 404	Lakeflow Connect for Monday.com
mst	PUBLIC_PREVIEW	✅ 200	Transactions
multiple_git_creds	GA	✅ 200	Multiple Git Credentials
netskope_logs_connector	BETA	🟡 404	Lakeflow Connect for Netskope Logs
new_policy_form	GA	✅ 200	New compute policy form
object_metadata_column	PUBLIC_PREVIEW	🟡 404	Object Metadata Column
oltp_database	GA	✅ 200	Lakebase Postgres
omnigents	BETA	🟡 404	Omnigent
one_chat	GA	✅ 200	New chat experience in Genie
operationalEmailCustomRecipient	GA	✅ 200	-
otel_collector	PUBLIC_PREVIEW	✅ 200	OpenTelemetry on Databricks
otel_model_serving	GA	✅ 200	OpenTelemetry for Databricks Model Serving
outlook_connector	BETA	✅ 200	Lakeflow Connect for Outlook
pat_autoscoping	BETA	🟡 404	Personal access tokens auto-scoping
pendo_connector	BETA	🟡 404	Lakeflow Connect for Pendo
pipeline_parameters	BETA	✅ 200	Spark Declarative Pipeline Parameters
pkg_repo_api_cluster	GA_SOON	✅ 200	Default Python package repositories in clusters created via API
pkg_repo_dlt	GA_SOON	✅ 200	Default Python package repositories in Spark Declarative Pipelines
pkg_repo_ui_cluster	GA_SOON	✅ 200	Default Python package repositories in clusters created via UI
power_bi_task	PUBLIC_PREVIEW	✅ 200	Power BI task type
query_perf_insights	BETA	✅ 200	Query performance insights
remote_ds_writes	GA	✅ 200	Remote data sources write support on serverless compute
remote_query_tvf	PUBLIC_PREVIEW	✅ 200	Enables remote query table-valued function (remote_query).
scoped_pat	GA	✅ 200	Scoped personal access tokens
serverless_jar_jobs	PUBLIC_PREVIEW	✅ 200	Serverless JARs
serverless_workload_observability	BETA	🟡 404	Improved Lakeflow Performance Observability
sf_mktg_connector	BETA	✅ 200	Lakeflow Connect for Salesforce Marketing Cloud
sfdc_file_sharing	GA	✅ 200	Salesforce Data Cloud file sharing federation
sftp_connector	GA	✅ 200	SFTP Connector
sharepoint_connector	BETA	✅ 200	Lakeflow Connect for Sharepoint
slack_ailogs_connector	BETA	🟡 404	Lakeflow Connect for Slack Access and Integration Logs
smartsheet_connector	BETA	✅ 200	Lakeflow Connect for Smartsheet
square_connector	BETA	🟡 404	Lakeflow Connect for Square
standalone_mv_st_on_serverless_gc	BETA	🟡 404	MV and ST in Serverless Notebooks and Jobs
supervisor_api	BETA	🟡 404	Supervisor API
system_managed_job	BETA	✅ 200	System-Managed Job for Materialized Views & Streaming Tables
tabular_subscription_attachments	GA	🟡 404	Widget Data Attachments for Dashboard Subscriptions
third_party_agent_connectors	BETA	🟡 404	Third Party Connectors for Agents
tiktok_ads_connector	BETA	✅ 200	Lakeflow Connect for TikTok Ads
tiles_platform	BETA	✅ 200	Mosaic AI Agent Bricks Preview
time_type	BETA	🟡 404	Time data type
tut_delta_sharing_ws	BETA	✅ 200	Table Update Triggers on OpenSharing (Recipient)
ucmr_prompt_registry	BETA	✅ 200	Managed MLflow Prompt Registry
variant_shredding	BETA	✅ 200	Variant Shredding for Optimized Read Performance on Semi-Structured Data
vector_search_rerank	GA	✅ 200	Vector Search Reranker
vectorsearch_highqps	PUBLIC_PREVIEW	✅ 200	Vector Search High QPS
vs_autoeval	BETA	🟡 404	AI Search: Quality Evaluation
vs_full_text	BETA	✅ 200	AI Search: Full-Text Search
wday_hcm_connector	BETA	✅ 200	Lakeflow Connect for Workday HCM
wh_activity_details	GA	✅ 200	Warehouse Activity Details
wiz_alogs_connector	BETA	🟡 404	Lakeflow Connect for Wiz Audit Logs
wsfs_git_cli	BETA	✅ 200	Git CLI support for Git folders
zdesk_supt_connector	GA	✅ 200	Lakeflow Connect for Zendesk Support
zerobus_ingest_core	GA	✅ 200	Lakeflow Connect Zerobus Ingest
zip_connector	BETA	🟡 404	Lakeflow Connect for Zip
zoho_books_connector	BETA	🟡 404	Lakeflow Connect for Zoho Books
zoom_logs_connector	BETA	🟡 404	Lakeflow Connect for Zoom Logs

Details

agent_monitoring

• Display name: Production Monitoring for MLflow
• Phase: BETA
• Status: ✅ 200

This beta enables monitoring of any Generative AI app or agent deployed outside of Databricks OR on Databricks using MLflow 3.0. It provides developers with tools to track both performance metrics (latency, request volume, errors) and quality metrics (accuracy, correctness, compliance), allowing them to detect drift or regressions via LLM-based evaluations on production traffic. Developers can also deep dive into individual requests for debugging and improvement purposes, and export real-world logs into evaluation sets to drive continuous enhancements of their Generative AI applications.

{"boolean_val": {"value": true}}

agents_obo

• Display name: Agent Framework: On-Behalf-Of-User Authorization
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

This feature enables on-behalf-of-user authentication for generative AI agents deployed to Model Serving via Mosaic AI Agent Framework. When you deploy an agent to Model Serving that performs on-behalf of end user access using Mosaic AI agent framework, the agent will be able to access Databricks resources using the identity of the agent invoker.

{"boolean_val": {"value": true}}

aha_connector

• Display name: Lakeflow Connect for Aha!
• Phase: BETA
• Status: 🟡 404

Ingest from Aha! using a simple and efficient connector

{"boolean_val": {"value": true}}

ai_classify

• Display name: AI Classify
• Phase: GA
• Status: ✅ 200

The ai_classify() function enables you to classify input text directly in SQL using state-of-the-art generative AI models provided by Databricks Foundation Model APIs. By supplying a set of labels, you can declaratively assign categories to unstructured text.

{"boolean_val": {"value": true}}

ai_extract

• Display name: AI Extract
• Phase: GA
• Status: ✅ 200

The ai_extract() function enables you to extract structured entities from unstructured text directly in SQL using state-of-the-art generative AI models provided by Databricks Foundation Model APIs. After specifying a target schema, you can declaratively transform raw text into structured outputs.

{"boolean_val": {"value": true}}

ai_parse_document

• Display name: AI ParseDocument
• Phase: GA
• Status: ✅ 200

The ai_parse_document() function invokes a state-of-the-art generative AI model from Databricks Foundation Model APIs to extract structured content from unstructured documents.

{"boolean_val": {"value": true}}

ai_prep_search

• Display name: AI Prep Search
• Phase: BETA
• Status: 🟡 404

The ai_prep_search() function enables you to transform the output of ai_parse_document into a format optimized for vector search and information retrieval systems.

{"boolean_val": {"value": true}}

ai_runtime_beta_features

• Display name: AI Runtime Beta Features
• Phase: BETA
• Status: 🟡 404

This preview allows users to use AI Runtime Beta features in their workspaces.

{"boolean_val": {"value": true}}

air_h100_multinode

• Display name: Serverless GPU Compute API Remote H100s
• Phase: BETA
• Status: ✅ 200

Enables users to use Serverless GPU Compute Python APIs to submit remote distributed training workloads using single and multi-node H100 GPUs for users of Serverless GPU Compute.

{"boolean_val": {"value": true}}

air_interactive

• Display name: Serverless GPU Compute
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Serverless GPU compute provides access to GPU compute (e.g A10 GPUs) via serverless. Users can, within the environment panel, select an accelerator to run notebook workloads like model training and finetuning without having to provision or manage infrastructure, streamlining development and training of models.

{"boolean_val": {"value": true}}

alerts_v2

• Display name: SQL Alerts V2
• Phase: GA
• Status: ✅ 200

The next generation of Databricks SQL Alerts - with a revamped user interface, a simplified data model, evaluation history, and improved API support.

{"boolean_val": {"value": true}}

alertv2_job_task

• Display name: Alert Job Task
• Phase: GA
• Status: ✅ 200

User can select Alert as a primary job task type.

{"boolean_val": {"value": true}}

allowedAppsUserApiScopes

• Display name: -
• Phase: GA
• Status: ✅ 200

Restrict which OAuth scopes app developers can request when acting on behalf of users. Specify a list of allowed scopes, or use "*" to allow all supported scopes. An empty list effectively disables user authorization for apps.

{"allowed_scopes": ["sample_string"]}

anomaly_detection_ws

• Display name: Data quality monitoring with anomaly detection (workspace level)
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

This feature allows you to be alerted on data quality anomalies (e.g. freshness, completeness) for your tables. By learning the behaviors of each table and intelligently setting thresholds, the feature allows you to easily alert on data quality incidents across all your important tables. Currently, the feature is limited to freshness monitoring and is a library you call in a notebook, but will soon also include completeness (row count) monitoring. The table health information will also be published in Unity Catalog so data consumers know if a table has an ongoing data quality incident.

{"boolean_val": {"value": true}}

apps_otel

• Display name: OpenTelemetry for Databricks Apps
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enables observability features for Databricks Apps.

{"boolean_val": {"value": true}}

apps_v2_ui

• Display name: Databricks Apps V2
• Phase: GA
• Status: 🟡 404

Enables the standalone Databricks Apps V2 UI experience with dedicated app management pages.

{"boolean_val": {"value": true}}

authoring_context

• Display name: Focused notebook & file editor for Git folders
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Similar to opening a folder in an IDE, you can now set the scope of the notebook and file editor to a specific Git folder. When the scope is set to a Git folder, the side panel displays that folder’s contents as an expandable tree, and the editor’s tab bar displays only the files, notebooks, and queries opened while the scope is set to that folder.

{"boolean_val": {"value": true}}

cld_to_volumes

• Display name: Cluster Log Delivery to UC Volumes
• Phase: GA
• Status: ✅ 200

Allows users to configure a classic compute cluster’s cluster log delivery to a UC volume path through UI or API. Cluster logs for the Spark driver node, worker nodes and events will be delivered to the configured volume path.

{"boolean_val": {"value": true}}

cloudfiles_excel

• Display name: Excel File Format Support
• Phase: GA
• Status: ✅ 200

Read Excel files using Spark batch and streaming APIs including Auto Loader, read_files, spark.read and COPY INTO. Available in DBR17.1+

{"boolean_val": {"value": true}}

collaboration_platform_connectivity

• Display name: -
• Phase: GA
• Status: ✅ 200

Controls which external collaboration platforms (Slack and/or Microsoft Teams) can connect to this workspace. Defaults to ALLOW_ALL.

{"connectivity": "CONNECTIVITY_UNSPECIFIED"}

collaboration_platform_message_visibility

• Display name: -
• Phase: GA
• Status: ✅ 200

When enabled, users can choose whether responses from the Databricks app in Slack are visible to the channel. When disabled, all responses are forced to be private to the requesting user.

{"boolean_val": {"value": true}}

confluence_connector

• Display name: Lakeflow Connect for Confluence
• Phase: GA
• Status: ✅ 200

Ingest from Confluence with a simple and efficient connector

{"boolean_val": {"value": true}}

conn_cdc_col_select

• Display name: Lakeflow Connect Column Selection for Database Sources
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enable column selection for Lakeflow Connects SQL Server connector. Available only via API.

{"boolean_val": {"value": true}}

custom_apps_preview

• Display name: Databricks Apps
• Phase: GA
• Status: ✅ 200

Databricks Apps is the fastest way for data developers to build secure and governed apps for internal use.

{"boolean_val": {"value": true}}

custom_llm_serving

• Display name: Custom LLM Serving for Databricks Model Serving
• Phase: BETA
• Status: ✅ 200

Enable serving custom LLMs in model serving endpoints.

{"boolean_val": {"value": true}}

customerApprovedWSLoginExpirationTime

• Display name: -
• Phase: GA
• Status: ✅ 200

Enable and disable access to your workspace to Databricks personnel. You can control when and how long Databricks can access your workspace for support purposes. If you have disabled access, no Databricks personnel will be able to gain access to your workspace. Accepted string values are - '', 'indefinite' or valid ISO date

{"string_val": {"value": "string"}}

dabs_templates

• Display name: Custom bundle templates in the workspace
• Phase: BETA
• Status: ✅ 200

This preview enables admins to set up custom templates that users can use to get started with asset bundles directly in the Workspace. It allows CI/CD stakeholders to enforce best practices and standardized formats, helping users skip repetitive setup and launch projects faster.

{"boolean_val": {"value": true}}

dabs_visual_edit

• Display name: Visual authoring: UI <> YAML Sync for DABs in the Workspace
• Phase: BETA
• Status: ✅ 200

This preview enables users to edit resources (Jobs, Pipelines) in the UI and have YAML automatically sync when using Declarative Automation Bundles in the Workspace. Users can set up their resources for CI/CD without explicit knowledge of asset bundles.

{"boolean_val": {"value": true}}

dashboard_authoring_agent

• Display name: Genie Code for dashboard authoring
• Phase: GA
• Status: 🟡 404

Enables Genie Code to create and edit AI/BI Dashboards by delegating complex tasks like visualization generation, filter configuration, and layout organization

{"boolean_val": {"value": true}}

databricks_one

• Display name: Databricks One
• Phase: GA
• Status: ✅ 200

Enables the Databricks One business user experience for Consumer entitled users in the Workspace.

{"boolean_val": {"value": true}}

dataclassificationws

• Display name: Data Classification
• Phase: GA
• Status: ✅ 200

This feature allows you to classify and review sensitive data across your entire catalog, powered by agents.

{"boolean_val": {"value": true}}

datascience_agent

• Display name: Genie Code
• Phase: GA
• Status: ✅ 200

Genie Code can automate multiple steps. From a single prompt, it can retrieve relevant assets, generate and run code, fix errors automatically, and visualize results. It can sample data and cell outputs to provide better results.

{"boolean_val": {"value": true}}

dbsql_5xl_pro

• Display name: 5XL PRO SQL Warehouse
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enable 5X-Large endpoint size for Pro SQL warehouses

{"boolean_val": {"value": true}}

dbsql_5xl_serverless

• Display name: 5XL Serverless SQL Warehouse
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enable 5X-Large endpoint size for Serverless SQL Warehouses

{"boolean_val": {"value": true}}

dbt_cloud_task

• Display name: dbt platform task
• Phase: BETA
• Status: ✅ 200

The dbt platform task lets you run your dbt platform jobs as part of existing databricks workflows

{"boolean_val": {"value": true}}

dc_custom_tags

• Display name: Extend Data Classification with Custom Classifiers
• Phase: BETA
• Status: 🟡 404

Automatically detect custom/business-specific tags beyond Data Classification’s supported set.

{"boolean_val": {"value": true}}

default_base_envs

• Display name: Workspace base environments for serverless compute
• Phase: GA
• Status: ✅ 200

Workspace base environments ensure that users automatically have a predefined environment version and set of Python packages available in their notebooks.

{"boolean_val": {"value": true}}

default_wh_setting

• Display name: Default warehouse setting
• Phase: GA
• Status: ✅ 200

The default warehouse feature allows Databricks workspace admins to define a default SQL warehouse that is pre-selected for users across SQL authoring surfaces, including Catalog Explorer, SQL Editor, Dashboards, Alerts, and Genie. Users can optionally override this default to set a customer default warehouse for themselves.

{"boolean_val": {"value": true}}

designer

• Display name: Lakeflow Designer
• Phase: GA
• Status: ✅ 200

Allows users to use Lakeflow Designer, a low-code, fully governed, AI-native tool for data transformations.

{"boolean_val": {"value": true}}

direct_cdc_connector

• Display name: LakeFlow Connect for Direct Cdc Managed Ingestion Pipeline
• Phase: BETA
• Status: ✅ 200

Ingest from several databases instances for database connectors like SQL Server, Postgres, Oracle using Direct Cdc Managed Ingestion Pipeline.

{"boolean_val": {"value": true}}

disable_legacy_access

• Display name: -
• Phase: GA
• Status: ✅ 200

'Disabling legacy access' has the following impacts: 1. Disables direct access to Hive Metastores from the workspace (access through Hive Metastore federation is unaffected). 2. Disables fallback mode on external location access from the workspace. 3. Disables Databricks Runtime versions prior to 13.3LTS. Important: this setting cannot be enabled while the workspace default catalog is set to a legacy catalog (hive_metastore or spark_catalog). Change the default catalog to a Unity Catalog catalog first.

{"boolean_val": {"value": true}}

disable_legacy_dbfs

• Display name: -
• Phase: GA
• Status: ✅ 200

Disabling legacy DBFS has the following two implications: 1. Access to DBFS root and DBFS mounts is disallowed (as well as the creation of new mounts). 2. Disables Databricks Runtime versions prior to 13.3LTS. When the setting is off, all DBFS functionality is enabled and no restrictions are imposed on Databricks Runtime versions. This setting can take up to 20 minutes to take effect and requires a manual restart of all-purpose compute clusters and SQL warehouses.

{"boolean_val": {"value": true}}

discover_page

• Display name: Discover Page
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enables the Discover Page in the left nav and the search empty state for this workspace. The Discover Page is a curated, user-friendly interface that helps users find and understand their organizations most important UC and workspace assets. Note: The account-level Domain and Discover flag must also be enabled for the Discover Page to appear in the local workspace.

{"boolean_val": {"value": true}}

ds_v2_join_pushdown

• Display name: Join Pushdown for Federated Queries
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enables automatic pushdown of join operations to remote databases when executing federated queries through Lakehouse Federation. When enabled, Databricks pushes inner, left, and right joins between tables from the same JDBC datasource (Oracle, PostgreSQL, MySQL, SQL Server, Teradata) directly to the remote database engine, reducing data transfer and improving query performance. This feature is available on Databricks Runtime 17.2 and above.

{"boolean_val": {"value": true}}

dynamics_connector

• Display name: Lakeflow Connect for Dynamics 365
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Ingest Dynamics 365 data via Dataverse with a simple and efficient connector. Available via API or UI.

{"boolean_val": {"value": true}}

embedded_genie

• Display name: Embed Genie as an iframe
• Phase: GA
• Status: 🟡 404

This feature allows Genie space authors to embed Genie spaces into external applications. Users in the external application must have access to Databricks and the Genie space.

{"boolean_val": {"value": true}}

enable_dcs_vnext

• Display name: DCS-vNext
• Phase: BETA
• Status: ✅ 200

Enables DCS-vNext support, which uses the LXC container and Dataplane Runtime daemon and Sandbox API to run customer containers.

{"boolean_val": {"value": true}}

enable_dlmv_aibi

• Display name: Enable Dashboard Local Metric Views in AI/BI Dashboards
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Create Metric Views that are local to an Ai/BI dashboard. These Dashboard Local Metric Views are created as datasets.

{"boolean_val": {"value": true}}

enable_github_webhook_app_deployments

• Display name: Databricks Apps - GitHub Webhook-Triggered Deployments
• Phase: BETA
• Status: 🟡 404

Automatically deploy apps when changes are pushed to a connected GitHub repository. Push events received via GitHub webhooks trigger app deployment workflows, removing the need for manual redeployment after code changes.

{"boolean_val": {"value": true}}

enable_lakeview_tags

• Display name: Tagging support for workspace scoped assets
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enable tags for new workspace scoped assets including lakeview dashboards and genie spaces. Users with edit level or above permissions can add and modify tags on the asset

{"boolean_val": {"value": true}}

enable_obo_user_apps

• Display name: Databricks Apps - On-Behalf-Of User Authorization
• Phase: GA_SOON
• Status: ✅ 200

Allows the Databricks App to act on behalf of the app user. This enhancement allows the app to honor the user's access permissions defined in Unity Catalog and in Databricks Workspace.

{"boolean_val": {"value": true}}

enforceGitAppDeployments

• Display name: -
• Phase: GA
• Status: ✅ 200

When enabled, apps in this workspace can only be deployed from Git.

{"boolean_val": {"value": true}}

excel_connector_flip

• Display name: Excel Add-In
• Phase: PUBLIC_PREVIEW
• Status: 🟡 404

Enable the Databricks Add-In for Excel to import data from Databricks into Excel

{"boolean_val": {"value": true}}

exp_sp_token_notif

• Display name: Expiring service principal access token notifications
• Phase: BETA
• Status: ✅ 200

Enable automatic email notifications for service principal owned expiring personal access tokens. When enabled, this feature proactively monitors service principal access tokens in your workspace and sends email notifications to workspace admins before their tokens expire, helping prevent service disruptions and authentication failures.

{"boolean_val": {"value": true}}

external_access_to_managed_delta

• Display name: External Access to Unity Catalog Managed Delta Table
• Phase: PUBLIC_PREVIEW
• Status: 🟡 404

External Access to Unity Catalog Managed Delta Tables lets external engines like Apache Spark, Trino, and other Delta clients securely create, read and write Unity Catalog-managed Delta tables using Databricks open APIs and short-lived, UC-governed credentials.

{"boolean_val": {"value": true}}

external_engine_fgac

• Display name: Cross-engine ABAC
• Phase: BETA
• Status: 🟡 404

This feature enables external clients to securely access tables in Databricks via UC Open APIs with attribute-based access controls (ABAC), row filters, and column masks enforced.

{"boolean_val": {"value": true}}

filebrowser_tree

• Display name: Tree view of the side panel file browser
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

The file browser in the editor’s side panel now supports a tree view, allowing you to expand multiple folders at once and quickly navigate between files across directories without backtracking.

{"boolean_val": {"value": true}}

fmapi_qwen3_instruct

• Display name: Enable Extended Models
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

This preview enables an extended family of models (e.g. Qwen) in model serving.

{"boolean_val": {"value": true}}

foreign_tbl_comments

• Display name: Comments on Foreign Tables
• Phase: BETA
• Status: ✅ 200

Allows fetching in comments from the foreign systems.

{"boolean_val": {"value": true}}

fstore_decl_fw

• Display name: Feature Views (Batch)
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Opt in to the new Feature Views for defining batch features in Databricks Feature Store. You can still use the legacy feature table pipelines alongside the new framework. Note: This preview currently supports batch features only; streaming features is available in a separate preview.

{"boolean_val": {"value": true}}

fstore_decl_strm_fw

• Display name: Feature Store Streaming Feature Views
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enables streaming Feature Views in Databrick's Feature Store.

{"boolean_val": {"value": true}}

full_screen_genie_code

• Display name: Full Page Genie Code
• Phase: BETA
• Status: 🟡 404

The "Full Page" mode and toggle of Genie Code. When enabled, customers have an option to open Genie Code in a Full Page UI with assets that can be opened in soft tabs.

{"boolean_val": {"value": true}}

full_text_search_index

• Display name: SQL: Full-Text Search Index for UC managed tables
• Phase: BETA
• Status: 🟡 404

Use the CREATE SEARCH INDEX SQL statement to create full-text search indexes on text columns of UC managed Delta Lake or Iceberg tables. The index accelerates SQL queries with selective, substring and word matching WHERE predicates.

{"boolean_val": {"value": true}}

gdrive_connector

• Display name: Lakeflow Connect for Google Drive
• Phase: BETA
• Status: ✅ 200

Ingest from Google Drive with a simple and efficient connector. Requires DBR 17.3+

{"boolean_val": {"value": true}}

generic_lfc

• Display name: Lakeflow Connect Community Connectors
• Phase: BETA
• Status: ✅ 200

This feature enables a wide range of OSS connectors built on the Python Data Source API and SDP, covering sources beyond managed ingestion and enabling users to build custom connectors.

{"boolean_val": {"value": true}}

genie_bi_migration

• Display name: Import from External BI to AI/BI
• Phase: BETA
• Status: 🟡 404

This preview allows users to import data models & dashboards from external BI tools as AI/BI Metric Views and Dashboards via Genie Code

{"boolean_val": {"value": true}}

genie_chat_sharing

• Display name: Genie Chat Sharing
• Phase: BETA
• Status: ✅ 200

Share Genie space chats with space managers by default, and allow users to share their chats with others in the Databricks account.

{"boolean_val": {"value": true}}

genie_deep_research

• Display name: Genie Agent
• Phase: GA
• Status: ✅ 200

The Genie Agent provides deeper data insights and answers complex business questions using multi-step reasoning and hypothesis investigation.

{"boolean_val": {"value": true}}

genie_inspect_answer

• Display name: Genie Answer Inspection
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Advanced technique that reviews initial SQL answers and makes improvements in standard Genie

{"boolean_val": {"value": true}}

genie_unstructured_files_in_deepresearch_mode

• Display name: Upload Local PDFs to Genie Spaces
• Phase: BETA
• Status: 🟡 404

Enables users to upload local PDFs as conversation-level context in Agent mode

{"boolean_val": {"value": true}}

github_connector

• Display name: Lakeflow Connect for Github
• Phase: BETA
• Status: ✅ 200

Ingest from Github with a simple and efficient connector.

{"boolean_val": {"value": true}}

google_ads_connector

• Display name: Lakeflow Connect for Google Ads
• Phase: BETA
• Status: ✅ 200

Ingest from Google Ads with a simple and efficient connector.

{"boolean_val": {"value": true}}

hspot_mktg_connector

• Display name: Lakeflow Connect for HubSpot
• Phase: GA
• Status: ✅ 200

Ingest from HubSpot with a simple and efficient connector.

{"boolean_val": {"value": true}}

icebergv3

• Display name: Iceberg V3
• Phase: GA
• Status: ✅ 200

Iceberg V3 adds performance and feature capabilities to Delta UniForm and Managed Iceberg tables. These features include deletion vectors for efficient row-level deletes, row lineage for incremental processing, and the Variant data type for semi-structured data.

{"boolean_val": {"value": true}}

ip_functions

• Display name: Ip Functions
• Phase: PUBLIC_PREVIEW
• Status: 🟡 404

Ip Functions feature preview enables built-in functions for working with IP addresses and CIDR blocks. Requires DBR 18.2 or later.

{"boolean_val": {"value": true}}

jdbc_connector

• Display name: Custom JDBC on UC Compute
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

This feature enables users to connect to data sources using a custom JDBC driver through the Spark Data Source API. The new UC Connection of type JDBC, runs a user-provided JDBC driver powered by Lakeguard isolation on UC-supported compute: serverless, standard, and dedicated clusters with DBR 18.1 or higher.

{"boolean_val": {"value": true}}

jdbc_oauth_m2m_connector

• Display name: OAuth M2M Support for Custom JDBC on UC Compute
• Phase: BETA
• Status: 🟡 404

This feature enables users to connect to data sources using a custom JDBC driver through the Spark Data Source API. The new Unity Catalog (UC) Connection of type JDBC utilizes OAuth M2M authentication and runs a user-provided JDBC driver powered by Lakeguard isolation on UC-supported compute: serverless, standard, and dedicated clusters with Databricks Runtime (DBR) 18.1 or higher.

{"boolean_val": {"value": true}}

jira_connector

• Display name: Lakeflow Connect for Jira
• Phase: BETA
• Status: ✅ 200

Ingest Jira data with a simple and efficient connector. Available via API for both Jira Cloud and on premise instances.

{"boolean_val": {"value": true}}

jobs_disabled_tasks

• Display name: Disabled tasks in Lakeflow Jobs
• Phase: GA
• Status: ✅ 200

Be able to disable tasks in Lakeflow Jobs. By disabling a task, Lakeflow Jobs will not run that task in subsequent runs until you re-enable it.

{"boolean_val": {"value": true}}

jobs_serverless_managed_base_environments

• Display name: Serverless workspace base environment support in Jobs
• Phase: BETA
• Status: 🟡 404

Enable serverless workspace base environment support for Notebook, Python Script, and Wheel tasks in Jobs

{"boolean_val": {"value": true}}

lakebase_accel_sync

• Display name: Lakebase Accelerated Sync
• Phase: PRIVATE_PREVIEW
• Status: ✅ 200

Uses parallelized load directly into object storage for faster initial load and refresh.

{"boolean_val": {"value": true}}

lakebase_cdf

• Display name: Lakebase CDF
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Lakebase-native export of the Postgres Change Data Feed to Delta tables. Supports scale to zero with Lakebase Autoscaling.

{"boolean_val": {"value": true}}

lakebase_otel_integration

• Display name: Lakebase OpenTelemetry Integration
• Phase: BETA
• Status: 🟡 404

Enables OpenTelemetry integration for Lakebase, allowing export of telemetry data to external observability platforms.

{"boolean_val": {"value": true}}

lakebase_search

• Display name: Lakebase Search
• Phase: BETA
• Status: 🟡 404

This preview introduces scalable vector search and native BM-25 fulltext search to Lakebase.

{"boolean_val": {"value": true}}

lakeflow_new_jobs_ui

• Display name: Lakeflow Jobs UI
• Phase: GA
• Status: ✅ 200

The new Lakeflow Jobs UI features a streamlined layout, redesigned task palette, and a cleaner empty state canvas for a faster, more focused experience. To enable it, open any job and switch on the toggle on the top of the page, you can opt-out at any time

{"boolean_val": {"value": true}}

lakeflow_qbc

• Display name: Lakeflow Connect Query Based Connectors
• Phase: GA
• Status: ✅ 200

Enable Lakeflow Connect query based connectors to ingest from Lakehouse federation sources.

{"boolean_val": {"value": true}}

lakeflow_runs_list

• Display name: Unified Runs List
• Phase: GA
• Status: ✅ 200

You can now view all of your Jobs and Pipeline executions in the updated Runs list. Track all of your pipeline executions in one place and filter by status, time, run-as user, and error codes in real time. Spot trends using the visualization and summary of the current top five error codes.

{"boolean_val": {"value": true}}

lakehouse_replay

• Display name: Lakehouse Replay
• Phase: BETA
• Status: ✅ 200

Databricks automatically replays read‑only serverless workloads in a safe shadow environment so regressions are detected and fixed before they impact production workloads.

{"boolean_val": {"value": true}}

lf_pipelines_auth

• Display name: Lakeflow Pipelines Editor
• Phase: GA
• Status: ✅ 200

Purpose-built IDE for declarative data pipelines. Designed to support everything you need for building pipelines in one place: code-first authoring, folder-based organization, selective execution, data previews, and pipeline graphs. Integrated with the Databricks Platform, supporting version control, code reviews, and scheduling.

{"boolean_val": {"value": true}}

llm_proxy_partner_powered

• Display name: -
• Phase: GA
• Status: ✅ 200

Determines if partner powered models are enabled or not. This setting is available at both account and workspace level

{"boolean_val": {"value": true}}

managed_mcp_servers

• Display name: Managed MCP Servers
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Databricks managed MCP (Model Context Protocol) servers enable your AI agents to access data and tools governed in Databricks, with data governance and permissions enforced out of the box

{"boolean_val": {"value": true}}

managed_memory_agents

• Display name: Managed Memory for Agents
• Phase: BETA
• Status: 🟡 404

Enable Managed Memory APIs and Memory Store Unity Catalog Securable for Custom Agents.

{"boolean_val": {"value": true}}

marketplace_app_install

• Display name: Marketplace - Install Databricks Apps
• Phase: PUBLIC_PREVIEW
• Status: 🟡 404

Allow workspace admins to install databricks apps from Databricks Marketplace.

{"boolean_val": {"value": true}}

meta_ads_connector

• Display name: Lakeflow Connect for Meta Ads
• Phase: BETA
• Status: ✅ 200

Ingest Meta Ads data with a simple and efficient connector.

{"boolean_val": {"value": true}}

mlflow_logged_models

• Display name: Models in Unity Catalog: Deployment Jobs
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Deployment jobs allow you to manage the model lifecycle by automating tasks like evaluation, approval, and deployment whenever a new model version is created, integrating seamlessly with Unity Catalog models and Databricks Jobs. These jobs simplify the setup of model deployment pipelines, incorporate human-in-the-loop approvals, and provide governed workflows with clear visibility into progress and historical context for each model version.

{"boolean_val": {"value": true}}

model_triggers

• Display name: Model update job triggers
• Phase: BETA
• Status: ✅ 200

Model update job triggers are a new type of job trigger that responds to UC model metadata events, such as a new model is created, or a new model version is created, or a model alias is set on a model version. Model update job triggers allow arbitrary jobs to be triggered based on the occurence of these events at varying scopes, including events happening under a specific model, a specific schema, or the whole metastore.

{"boolean_val": {"value": true}}

monday_com_connector

• Display name: Lakeflow Connect for Monday.com
• Phase: BETA
• Status: 🟡 404

Ingest from Monday.com with a simple and efficient connector

{"boolean_val": {"value": true}}

mst

• Display name: Transactions
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Run multiple SQL statements across multiple Delta and Iceberg tables as a single, atomic transaction with full ACID guarantees. All changes succeed or roll back together, ensuring data consistency across your operations.

{"boolean_val": {"value": true}}

multiple_git_creds

• Display name: Multiple Git Credentials
• Phase: GA
• Status: ✅ 200

Ability to create and use multiple Git credentials per user for Git folders

{"boolean_val": {"value": true}}

netskope_logs_connector

• Display name: Lakeflow Connect for Netskope Logs
• Phase: BETA
• Status: 🟡 404

Ingest Netskope Logs with a simple and efficient connector

{"boolean_val": {"value": true}}

new_policy_form

• Display name: New compute policy form
• Phase: GA
• Status: ✅ 200

This preview enables a new and improved user interface for creating, editing and viewing compute policies. Non-admin users are now able to view compute policies they are assigned to.

{"boolean_val": {"value": true}}

object_metadata_column

• Display name: Object Metadata Column
• Phase: PUBLIC_PREVIEW
• Status: 🟡 404

Introduces the _object_metadata hidden column that exposes cloud object-level properties for each file read by a file-based data source.

{"boolean_val": {"value": true}}

oltp_database

• Display name: Lakebase Postgres
• Phase: GA
• Status: ✅ 200

A new Postgres compute type built for low-latency reads and writes. Autoscaling offers autoscaling compute, branching, instance restore, and more, while Provisioned provides high availability (HA) and integration with Databricks Apps.

{"boolean_val": {"value": true}}

omnigents

• Display name: Omnigent
• Phase: BETA
• Status: 🟡 404

Gates the Omnigent product.

{"boolean_val": {"value": true}}

one_chat

• Display name: New chat experience in Genie
• Phase: GA
• Status: ✅ 200

Chat with Databricks agents and third-party data sources from a single conversation in Genie, governed by Unity Catalog.

{"boolean_val": {"value": true}}

operationalEmailCustomRecipient

• Display name: -
• Phase: GA
• Status: ✅ 200

Additional recipient for this workspace's operational emails. When set, Databricks delivers these emails to both the workspace admins and this address. Available on AWS and GCP.

{"email": "sample_string"}

otel_collector

• Display name: OpenTelemetry on Databricks
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Enables ingestion of OpenTelemetry data into Unity Catalog managed Delta tables for MLflow Tracing.

{"boolean_val": {"value": true}}

otel_model_serving

• Display name: OpenTelemetry for Databricks Model Serving
• Phase: GA
• Status: ✅ 200

OpenTelemetry log/span/metric persistence for model serving endpoints.

{"boolean_val": {"value": true}}

outlook_connector

• Display name: Lakeflow Connect for Outlook
• Phase: BETA
• Status: ✅ 200

Ingest from Outlook with a simple and efficient connector.

{"boolean_val": {"value": true}}

pat_autoscoping

• Display name: Personal access tokens auto-scoping
• Phase: BETA
• Status: 🟡 404

Auto-scoping narrows token permissions to match actual API usage, which reduces the risk of over-privileged tokens. New tokens with a lifetime longer than 30 days have auto-scoping on by default. Owners can turn it off or select 'all-apis' at creation. After 30 days, enabled tokens are scoped based on observed API usage. Existing tokens with 'all-apis' scope are backfilled based on historical usage. Admins and token owners are notified and can review and modify auto-assigned scopes at any time.

{"boolean_val": {"value": true}}

pendo_connector

• Display name: Lakeflow Connect for Pendo
• Phase: BETA
• Status: 🟡 404

Ingest from Pendo using a simple and efficient connector

{"boolean_val": {"value": true}}

pipeline_parameters

• Display name: Spark Declarative Pipeline Parameters
• Phase: BETA
• Status: ✅ 200

Write extensible, maintainable pipeline code by parameterizing Spark Declarative Pipelines.

{"boolean_val": {"value": true}}

pkg_repo_api_cluster

• Display name: Default Python package repositories in clusters created via API
• Phase: GA_SOON
• Status: ✅ 200

Apply the default Python package repositories in clusters created via API. The changes of configurations apply to newly created clusters and existing clusters upon restart.

{"boolean_val": {"value": true}}

pkg_repo_dlt

• Display name: Default Python package repositories in Spark Declarative Pipelines
• Phase: GA_SOON
• Status: ✅ 200

Default Python package repositories in serverless and classic Spark Declarative Pipelines (SDP). The changes of configurations apply to new pipeline runs.

{"boolean_val": {"value": true}}

pkg_repo_ui_cluster

• Display name: Default Python package repositories in clusters created via UI
• Phase: GA_SOON
• Status: ✅ 200

Apply the default Python package repositories in clusters created via UI. The changes of configurations apply to newly created clusters and existing clusters upon restart.

{"boolean_val": {"value": true}}

power_bi_task

• Display name: Power BI task type
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

The Power BI task type in Databricks Workflows allows users to keep Power BI semantic models up-to-date with source data in Unity Catalog. The task can sync table metadata, evolving schemas, and primary/foreign key relationships. It works for DirectQuery and Import mode at the table level, allowing you to define composite models.

{"boolean_val": {"value": true}}

query_perf_insights

• Display name: Query performance insights
• Phase: BETA
• Status: ✅ 200

When queries run, Databricks might return insights that identify opportunities to improve performance.

{"boolean_val": {"value": true}}

remote_ds_writes

• Display name: Remote data sources write support on serverless compute
• Phase: GA
• Status: ✅ 200

Enables the write support on serverless compute for the remote data sources. The feature works only for writes through the DataFrame API. This feature doesnt include Lakehouse Federation writes.

{"boolean_val": {"value": true}}

remote_query_tvf

• Display name: Enables remote query table-valued function (remote_query).
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Function allows users to execute query in remote engine syntax using credentials from a Unity Catalog connection. Function is available on Databricks Runtime 17.3 or above.

{"boolean_val": {"value": true}}

scoped_pat

• Display name: Scoped personal access tokens
• Phase: GA
• Status: ✅ 200

Scoped personal access tokens (Scoped PATs) limit which APIs each token can call through specific scopes that define its permissions. The feature helps enforce least-privilege access and reduces risk if a token is compromised. You can configure scopes individually on each token and update them at any time to match the token's intended use—such as connecting BI tools, or running automation scripts.

{"boolean_val": {"value": true}}

serverless_jar_jobs

• Display name: Serverless JARs
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Deploy Scala and Java jobs as Jars on serverless compute with faster startup, auto-scaling, and no cluster management

{"boolean_val": {"value": true}}

serverless_workload_observability

• Display name: Improved Lakeflow Performance Observability
• Phase: BETA
• Status: 🟡 404

Improved Lakeflow Performance Observability includes aggregates of query metrics and insights on the workload level, support for timeline view for single task runs and other improvements.

{"boolean_val": {"value": true}}

sf_mktg_connector

• Display name: Lakeflow Connect for Salesforce Marketing Cloud
• Phase: BETA
• Status: ✅ 200

Ingest from Salesforce Marketing Cloud with a simple and efficient connector.

{"boolean_val": {"value": true}}

sfdc_file_sharing

• Display name: Salesforce Data Cloud file sharing federation
• Phase: GA
• Status: ✅ 200

Query data from Salesforce Data Cloud using the zero-copy direct file access approach.

{"boolean_val": {"value": true}}

sftp_connector

• Display name: SFTP Connector
• Phase: GA
• Status: ✅ 200

Ingest files from SFTP server using Auto Loader. Requires DBR 17.3+.

{"boolean_val": {"value": true}}

sharepoint_connector

• Display name: Lakeflow Connect for Sharepoint
• Phase: BETA
• Status: ✅ 200

Ingest Sharepoint data with a simple and efficient connector. Available via API.

{"boolean_val": {"value": true}}

slack_ailogs_connector

• Display name: Lakeflow Connect for Slack Access and Integration Logs
• Phase: BETA
• Status: 🟡 404

Ingest Slack Access and Integration Logs with a simple and efficient connector

{"boolean_val": {"value": true}}

smartsheet_connector

• Display name: Lakeflow Connect for Smartsheet
• Phase: BETA
• Status: ✅ 200

Ingest from Smartsheet with a simple and efficient connector.

{"boolean_val": {"value": true}}

square_connector

• Display name: Lakeflow Connect for Square
• Phase: BETA
• Status: 🟡 404

Ingest from Square with a simple and efficient connector

{"boolean_val": {"value": true}}

standalone_mv_st_on_serverless_gc

• Display name: MV and ST in Serverless Notebooks and Jobs
• Phase: BETA
• Status: 🟡 404

Feature preview to enable creating and refreshing SDP Materialized Views and Streaming Tables in Serverless Notebooks and Jobs using Serverless Generic Compute.

{"boolean_val": {"value": true}}

supervisor_api

• Display name: Supervisor API
• Phase: BETA
• Status: 🟡 404

The Supervisor API lets you build custom agents on Databricks with a single request. You specify the model, tools, and instructions, and Databricks runs the agent loop for you by calling the model, executing tools, and producing the final response, with support for background mode for long-running tasks.

{"boolean_val": {"value": true}}

system_managed_job

• Display name: System-Managed Job for Materialized Views & Streaming Tables
• Phase: BETA
• Status: ✅ 200

Materialized View (MV) and Streaming Table (ST) schedules now surface as a system-managed refresh job, providing improved observability and configurability. Users can view the refresh job, configure schedules and trigger frequency, set notifications for refresh runs, and choose between standard or performance-optimized execution modes directly in the Catalog Explorer.

{"boolean_val": {"value": true}}

tabular_subscription_attachments

• Display name: Widget Data Attachments for Dashboard Subscriptions
• Phase: GA
• Status: 🟡 404

Enables support for attaching widget data to dashboard subscription email notifications. Widget data can be attached as CSV, TSV, or Excel.

{"boolean_val": {"value": true}}

third_party_agent_connectors

• Display name: Third Party Connectors for Agents
• Phase: BETA
• Status: 🟡 404

Enables Databricks AI features to use Databricks-managed connectors to popular third party providers.

{"boolean_val": {"value": true}}

tiktok_ads_connector

• Display name: Lakeflow Connect for TikTok Ads
• Phase: BETA
• Status: ✅ 200

Ingest from TikTok Ads with a simple and efficient connector.

{"boolean_val": {"value": true}}

tiles_platform

• Display name: Mosaic AI Agent Bricks Preview
• Phase: BETA
• Status: ✅ 200

Mosaic AI Agent Bricks enables you to build and optimize domain-specific agent systems. Simply select your problem, and Agent Bricks will run optimizations to improve quality and cost, generating a deployable endpoint on Databricks. Additionally, Agent Bricks provides automated evaluation and recommendations to refine your agent systems performance. Enabling this flag will also enable ai_parse_document function; however, for the regions where Agent Bricks is not available, you will only get access to ai_parse_document without Agent Bricks enabled.

{"boolean_val": {"value": true}}

time_type

• Display name: Time data type
• Phase: BETA
• Status: 🟡 404

Adds the support for TIME data type to Spark. Applies to DBR 18.2+ on all SQL Warehouses, Serverless GC and Standard compute.

{"boolean_val": {"value": true}}

tut_delta_sharing_ws

• Display name: Table Update Triggers on OpenSharing (Recipient)
• Phase: BETA
• Status: ✅ 200

Recipient workspace-level preview for Table Update Triggers on OpenSharing tables. When enabled for a workspace, users in that workspace can create Table Update Triggers (TUTs) on tables shared with them through OpenSharing. Requires the provider account to have the tut_delta_sharing preview enabled.

{"boolean_val": {"value": true}}

ucmr_prompt_registry

• Display name: Managed MLflow Prompt Registry
• Phase: BETA
• Status: ✅ 200

Managed MLflow Prompt Registry on Databricks is a powerful tool that streamlines prompt engineering and management in your Generative AI (GenAI) applications. It enables you to version, track, and reuse prompts across your organization, helping maintain consistency and improving collaboration in prompt development. The Prompt Registry also enables you to optimize prompts through a native integration with DSPy, delivering higher quality while minimizing time spent on manual prompt engineering.

{"boolean_val": {"value": true}}

variant_shredding

• Display name: Variant Shredding for Optimized Read Performance on Semi-Structured Data
• Phase: BETA
• Status: ✅ 200

Variant shredding extracts commonly occurring fields in semi-structured data into separate columns during writes. This significantly improves read performance, but incurs some overhead on writes. Variant shredding is supported in DBR 17.2 and above.

{"boolean_val": {"value": true}}

vector_search_rerank

• Display name: Vector Search Reranker
• Phase: GA
• Status: ✅ 200

When enabled, users can improve similarity search quality from Vector Search by reranking retrieved results with a specialized model. No provisioning or endpoint management required. Use the familiar similarity search APIs, and get better quality results.

{"boolean_val": {"value": true}}

vectorsearch_highqps

• Display name: Vector Search High QPS
• Phase: PUBLIC_PREVIEW
• Status: ✅ 200

Vector Search High QPS enables significantly higher real-time query throughput for Vector Search by scaling endpoint capacity via replication – so you can serve more concurrent search requests without standing up or load‑balancing across multiple endpoints. In this phase, you manually increase QPS capacity by adjusting endpoint replication and monitor performance using available metrics; support for automatic scaling is planned for later releases.

{"boolean_val": {"value": true}}

vs_autoeval

• Display name: AI Search: Quality Evaluation
• Phase: BETA
• Status: 🟡 404

AI Search Quality Evaluation enables automatic search quality evaluation for AI Search indexes, using synthetic query generation and LLM-based relevance scoring to measure recall, precision, NDCG, and other retrieval metrics.

{"boolean_val": {"value": true}}

vs_full_text

• Display name: AI Search: Full-Text Search
• Phase: BETA
• Status: ✅ 200

Users can create pure full-text indexes and run full-text search queries directly through the UI, Python SDK, or REST APIs. This bypasses ANN or vector search entirely and is especially valuable for use cases that require precise keyword matching.

{"boolean_val": {"value": true}}

wday_hcm_connector

• Display name: Lakeflow Connect for Workday HCM
• Phase: BETA
• Status: ✅ 200

Ingest from Workday HCM with a simple and efficient connector. To ingest from Workday RaaS, use the Workday Reports connector instead.

{"boolean_val": {"value": true}}

wh_activity_details

• Display name: Warehouse Activity Details
• Phase: GA
• Status: ✅ 200

Provides deeper visibility into SQL Warehouse usage by showing why a warehouse is running even when no queries are visible. The SQL Warehouse monitoring UI includes an activity details view in the running clusters chart, showing query execution and client-driven activity, such as open sessions or query fetching.

{"boolean_val": {"value": true}}

wiz_alogs_connector

• Display name: Lakeflow Connect for Wiz Audit Logs
• Phase: BETA
• Status: 🟡 404

Ingest audit logs from Wiz using a simple and efficient connector

{"boolean_val": {"value": true}}

wsfs_git_cli

• Display name: Git CLI support for Git folders
• Phase: BETA
• Status: ✅ 200

You can now run Git CLI commands against Git folders in the Web Terminal and from notebooks. Previously, there was only support for Git operations via the UI.

{"boolean_val": {"value": true}}

zdesk_supt_connector

• Display name: Lakeflow Connect for Zendesk Support
• Phase: GA
• Status: ✅ 200

Ingest from Zendesk Support with a simple and efficient connector.

{"boolean_val": {"value": true}}

zerobus_ingest_core

• Display name: Lakeflow Connect Zerobus Ingest
• Phase: GA
• Status: ✅ 200

Zerobus Ingest, part of Lakeflow Connect, is a robust API that allows you to efficiently push data into tables in a streaming, record-by-record, fashion, operating in a serverless multi-tenant environment to support a high volume of clients.

{"boolean_val": {"value": true}}

zip_connector

• Display name: Lakeflow Connect for Zip
• Phase: BETA
• Status: 🟡 404

Ingest from Zip with a simple and efficient connector

{"boolean_val": {"value": true}}

zoho_books_connector

• Display name: Lakeflow Connect for Zoho Books
• Phase: BETA
• Status: 🟡 404

Ingest from Zoho Books with a simple and efficient connector

{"boolean_val": {"value": true}}

zoom_logs_connector

• Display name: Lakeflow Connect for Zoom Logs
• Phase: BETA
• Status: 🟡 404

Ingest Zoom Logs with a simple and efficient connector

{"boolean_val": {"value": true}}