You can now runglobal queries, which let youreference data stored in more than one region in a single query. This feature isinPreview.

Control of MCP use with organization policies is deprecated. AfterMarch 17, 2026, organization policies that use thegcp.managed.allowedMCPServices constraint won't work, and you can controlMCP use with IAM deny policies. For more information about controlling MCP use,seeControl MCP use with IAM deny policies.

After March 17, 2026, when you enable BigQuery, the BigQuery MCP server isautomatically enabled.

You can now usedataset insightsto understand relationships between tables in a dataset by generatingrelationship graphs and cross-table queries. You can automatically generatedataset summaries, infer relationships across tables, and receive suggestionsfor analytical questions. This feature is inPreview.

You can now provide descriptions for the fields in your custom output schemawhen you use theAI.GENERATEandAI.GENERATE_TABLEfunctions.This feature isgenerally available(GA).

TheAI.CLASSIFY functionnow supports classifying your input into multiple categories. This feature is inPreview.

You can now run pipelines with three distinct execution methods: running alltasks, running selected tasks, and running tasks with selected tags. For moreinformation, seeRun a pipeline.This feature isgenerally available(GA).

You can now customize the scope of data documentation scans for BigQuery tablesto generate specific insights. You can choose to generate only SQL queries,only table and column descriptions, or all insights.

You can also create one-time data scans that execute immediately upon creation,removing the need for a separaterun command. These scans support aTime to Live (TTL) setting to automatically delete the scan resource aftercompletion.

For more information, seeGenerate insights for a BigQuery table.

Data transfers from theYouTube ChannelandYouTube Content Ownerdata sources now support reach reports. For more information, seeYouTube Channel report transformationandYouTube Content Owner report transformation.

You can now associatedata policies directly oncolumns. Thisfeature enables direct database administration for controlling access andapplying masking and transformation rules at the column level. This feature isnowgenerallyavailable (GA).

Gemini in BigQuery now processes data in the same jurisdiction (US orEU) asyour BigQuery datasets, or based upon user-specified location settings. For moreinformation, seeWhere Gemini BigQuery processes yourdata.

You can now passparameterized queriesfrom the BigQuery query editor in the Google Cloud console.

This feature isgenerally available(GA).

BigQuery now supports aRANDOM_HASH predefined masking rule. This rule returnsa hash of the column's value using a salted hash algorithm, and it providesstronger security than the standardHash (SHA-256) rule.

For more information, seeData masking rules.

BigQuery now offersconversational analytics,which accelerates data analysis by enabling insights through natural language.Users can view a predefined sample agent, chat with their BigQuery data orcustom agents, and access those agents even outside of BigQuery. They can alsousesupported BigQuery ML functionsin verified queries and in chat. This feature is inPreview.

You can now createBigQuery ML models by using theGoogle Cloud console.

This feature isgenerally available(GA).

The BigQuery change data capture feature has been renamed toBigQuery change data capture ingestion.

The BigQuery Data Transfer Service can nowtransfer data from Shopify to BigQuery.This feature is inPreview.

An updated version of theSimba JDBC driver for BigQueryis now available.

The BigQuery Data Transfer Service can nowtransfer data from Mailchimp to BigQuery.This feature is inPreview.

You can now use Gemini Cloud Assist todiscover resourcesacross your projects. For example, you can ask about a specific table's schema,or which tables contain demographic information about new users. This feature isinPreview.

You can now optionally specify which model to use by passing an endpointargument to theAI.IF,AI.SCORE,andAI.CLASSIFYfunctions.

You can now run queries that use theAI.IF,AI.SCORE,andAI.CLASSIFYfunctions by using yourend-user credentials instead of aBigQuery connection.

Support fortable parameters in table-valued functionsis restored.

BigQuery is now available in theBangkok (asia-southeast3) region.

You can now use Gemini Cloud Assist toget information about your job history,such as why a particular query was slow or which queries were the mostresource-intensive in the past day. This feature is inPreview.

Dataform workflows,BigQuery notebooks,pipelines,anddata preparationsare enforcing strict act-as mode at the project level. To avoid failures andmaintain automatic releases, you must use custom service accounts instead of thedefault Dataform service agent across all repositories. You must also grant theService Account User role (roles/iam.serviceAccountUser) to the defaultDataform service agent and relevant principals. For more information and toverify act-as permissions, seeUse strict act-as mode.

You can now use the Google-developed, open sourceJava Database Connectivity (JDBC) driver for BigQueryto connect your Java applications to BigQuery. This feature is inPreview.

TheCREATE EXTERNAL TABLEandLOAD DATAstatements now support the following options:

• time_zone: specify a time zone to use when loading data
• date_format,datetime_format,time_format, andtimestamp_format: define how date and time values are formatted in your source files
• null_markers: define the strings that representNULL values in CSV files.
• source_column_match: specify how loaded columns are matched to the schema. You can match columns by position or by name.

These features aregenerallyavailable (GA).

The BigQuery Data Transfer Service can nowtransfer data from PostgreSQL toBigQuery. This feature isgenerallyavailable (GA).

The BigQuery Data Transfer Service can nowtransfer data from Microsoft SQLServer to BigQuery. This feature is inPreview.

The BigQuery Data Transfer Service can nowtransfer data from MySQL toBigQuery. This feature isgenerallyavailable (GA).

You can now use the BigQuery Data Transfer Serviceto transfer data from blobstorage sources, such as Amazon SimpleStorage Service (Amazon S3), Azure Blob Storage, and Cloud Storage, into BigLakeIceberg tables in BigQuery. This feature is inPreview.

The BigQuery Data Transfer Service can now transfer data from the following datasources to BigQuery:

• Klaviyo
• HubSpot

These features are inPreview.

The BigQuery Data Transfer Service can nowtransfer data from Oracle toBigQuery. This feature isgenerally available (GA).

You can now use theBigQuery remote MCP serverto enable LLM agents to perform a range of data-related tasks.

This feature is inPreview.

An updated version of theODBC driver for BigQueryis now available.

You can now enableautonomous embedding generationon tables that you make with theCREATE TABLE statement.When you do this, BigQuery maintains a column of embeddings onthe table based on a source column. When you add or modify data in the sourcecolumn, BigQuery automatically generates or updates the embeddingcolumn for that data.

You can also use theAI.SEARCHfunction, enabling semantic search on tables that have autonomous embeddinggeneration enabled.

These features are inPreview.

Search results in theExplorerpane in BigQuery Studio now showresults in the current organization. You can use a drop-down menu to switchbetween organizations. This feature isgenerallyavailable (GA).

The BigQuery Data Transfer Service now supportsincremental data transferswhen transferring data from Salesforce to BigQuery. This feature is supported inPreview.

An updated version of theJDBC driver for BigQueryis now available.

You canset the default project and dataset for your pipelinein theSQLX options section, which simplifies task configuration by usingthese defaults for all tasks. This feature isgenerally available(GA).

You can now use the BigQuery Agent Analytics plugin within the Agent DevelopmentKit to export agent interaction data directly into BigQuery. This plugincaptures comprehensive logs of your agent's prompts, tool usage, and responses,enabling you to analyze and visualize agent performance metrics. The pluginleverages the BigQuery Storage Write API for efficient high-throughputstreaming. For more information on how to leverage this plugin in your agent,see theAnnouncing BigQuery Agent Analytics for the Google ADK.

You can now use Gemini in BigQuery tofix and explain errorsin your SQL queries. This feature is inPreview.

You can use theJSON_FLATTEN functionto extract all non-array values that are either directly in the inputJSONvalue or children of one or more consecutively nested arrays in the inputJSON value. This function is available inPreview.

Dataform now lets you automate the creation ofBigLake tables for Apache Iceberg in BigQuery.This feature isgenerally available(GA).

You can now useGemini 3.0when you call generative AI functions in BigQuery,such asAI.GENERATE.You must use the full global endpoint argument:

https://aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/global/publishers/google/models/gemini-3-pro-preview.

BigQuery ML now supports the followinggenerative AI functions:

• AI.GENERATE:generate free text to accomplish a wide range of tasks, such astranslation, summarization, and classification, on any unstructured data,including images, audio, video, and documents. It can also perform entityextraction and generate structured output. This function isgenerally available(GA).
• AI.EMBED:turn text, image, audio, video, or documents into embeddings. This functionis inPreview.
• AI.SIMILARITY:compute the semantic similarity between pairs of text, pairs of images, oracross text and images. This function is inPreview.
• You can use theAI.GENERATE_BOOL,AI.GENERATE_DOUBLE,andAI.GENERATE_INTfunctions to generate scalar values, which are convenient for filtering,scoring, and counting purposes.
• Each of these functions supportsauthentication with end-user credentials (EUC)to set up the necessary Vertex AI permissions.

BigQuery ML now supports the following table-valued generative AI functions:

• AI.GENERATE_TABLE:generate a table of structured output from unstructured data includingtext, images, audio, and video.
• AI.GENERATE_TEXTis the new, preferred version ofML.GENERATE_TEXT, which has the samefunctionality but with simplified column output names.
• AI.GENERATE_EMBEDDINGis the new, preferred version ofML.GENERATE_EMBEDDING, which has the samefunctionality but with simplified column output names.
• These functions are allgenerally available(GA).

You can nowpublish data insights,including query recommendations andauto-generated table and column descriptions, to the Dataplex Universal Catalog.This feature is inPreview.

You can usefolders to organize and controlaccess to single file code assets, such as notebooks, saved queries, datacanvases, and data preparation files. This feature is inPreview.

In the query execution graph, you can now use thequery textheatmap to identifywhich query text contributes to stages that consume more slot time, and to seequery plan details for those stages. This feature is inPreview.

You can nowshare SQL stored procedures in BigQuery sharing listingsand enablerole-based authorization for stored procedures.These features are inpreview.

The BigQueryOverview pageis now your hub for discovering tutorials, features, and resources to help youget the most out of BigQuery. It provides guided paths for usersof all skill levels. This feature is inPreview.

You can now use theinteractive SQL translator, thetranslation API, and thebatch SQL translator to translate thefollowing SQL dialects into GoogleSQL:

• Apache Impala SQL
• GoogleSQL (BigQuery)

Impala SQL translation can be used to migrate Cloudera and Apache Hadoop SQLworkloads that use Impala as a query engine.

GoogleSQL (BigQuery) translation can be used to verify and iteratively customizeyour translated SQL queries after an initial translation from an externaldialect. For example, you can apply systematic query rewrites usingYAMLconfigurationsto customize and optimize your GoogleSQL queries before deploying it.

These features are inPreview.

You can now usecustom constraints with anOrganization Policy to provide more granular control over specific fields forBigQuery dataset resources. This feature isgenerally available (GA).

You canaggregateanddeduplicatetable data with Gemini assistance in yourBigQuery data preparations.These features aregenerally available (GA).

Partitioning is now available forBigLake tables for Apache Iceberg in BigQuery.This feature is inPreview.

BigQuery ML now supports theTimesFM 2.5time series foundational model.You can use theTimesFM 2.5 model in theAI.FORECAST,AI.EVALUATE,andAI.DETECT_ANOMALIESfunctions to achieve better forecasting accuracy and lower latency.

BigQuery ML now offers theAI.DETECT_ANOMALIES function.Use theAI.DETECT_ANOMALIES function with a TimesFM model todetect anomaliesin time series data, using historical data as a baseline.This feature is inPreview.

The research paperARIMA_PLUS: Large-scale, Accurate, Automatic andInterpretable In-Database Time Series Forecasting and Anomaly Detection inGoogle BigQuery is now publicly available.This paper describes the algorithms behind theARIMA_PLUSandARIMA_PLUS_XREGmodels for time series forecasting and anomaly detection, and demonstrates thehigh performance, scalability, explainability, and customizability of themodels.

You can use theMATCH_RECOGNIZE clausein your SQL queries to filter and aggregate matches across rows in a table.This feature isgenerally available(GA).

You can now generatedata insights when youcreate aDataScanusing the Dataplex API. This feature isgenerally available(GA).

You can nowgenerate table and column descriptionsin all supported Gemini languages when you generate data insights.This feature isgenerally available(GA).

TheBigQuery Data Transfer Service for Google Ads now supportsGoogle Ads API v21.

You can now usecustom organization policies with the BigQuery migrationservice to allow or deny specificoperations during a BigQuery migration to meet your organization's complianceand security requirements. This includes an option to disable AI suggestionsduring a migration. This feature is inPreview.

We haveincreased the row capacityfor pivot tables backed by BigQuery inConnected Sheetsfrom 100,000 to 200,000 rows.

TheApache Iceberg REST catalog in BigLake metastoreis nowgenerally available(GA) with several new features, including BigQuery catalog federation,credential vending, and catalog management in the Google Cloud console.

You can nowgroupreservationstogether to prioritize idle slot sharing within the group. Reservations within areservation group share idle slots with each other before making them availableto other reservations in the project, giving you more control over slotallocation for high-priority workloads. This feature is inPreview.

Subscriber email logging lets you log the principal identifiers of userswho execute jobs and queries against linked datasets. You can enablelogging at thelisting leveland thedata exchange level.The logged data is available in thejob_principal_subject field of theINFORMATION_SCHEMA.SHARED_DATASET_USAGE view.This feature isgenerally available.

The BigQuery Data Transfer Service can now transfer data from thefollowing data sources:

• Facebook Ads
• Salesforce
• Salesforce Marketing Cloud
• ServiceNow

Transfers from these data sources are nowgenerally available(GA).

The administrative jobs explorer now includes ajob detailspage to help you diagnoseand troubleshoot queries. ThePerformance tab compiles query informationincluding the execution graph, SQL text, execution history, performancevariance, and system load during execution. You can alsocompare twojobs to identify discrepanciesand potential areas to improve query performance.

This feature is inPreview.

You can nowuse the Data Engineering Agentto use Gemini in BigQuery to build and modify datapipelines to ingest data into BigQuery. This feature is inpreview.

BigQuery now offers the followingmanaged AI functionsthat use Gemini to help you filter, join, rank, and classify your data:

• AI.IF:Filter and join text or multimodal data based on a condition described in natural language.
• AI.SCORE:Rate text or multimodal input to rank your data by quality, similarity, or other criteria.
• AI.CLASSIFY:Classify text into user-defined categories.

These functions are inPreview.

You can now use the Apache Arrow format tostream data to BigQuery with the Storage Write API.This feature isgenerally available(GA).

BigQuery is now offeringearly accessto conversational analytics. Conversational analytics accelerates data analysisby enabling quick insights through natural language. Users can chat with theirBigQuery data, create custom agents, and access those agents even outside ofBigQuery. To enroll in conversational analytics early access, fill out therequest form.

You can now use custom constraints with Organization Policy to provide moregranular control over specific fields for some BigQuery sharingresources. For more information, seeManage Sharing data exchanges and listings using custom constraints.This feature is inpreview.

Support fortable parameters in table-value functions (TVFs)has been temporarily disabled. We are working to restore this feature as soonas possible.

BigQuery ML now offers a built-inTimesFM univariate time series forecasting modelthat implements Google Research's open source TimesFM model. You can useBigQuery ML's built-in TimesFM model with the following functions:

• UseAI.FORECASTto perform forecasting. This function now supports a larger context window.
• UseAI.EVALUATEto evaluate forecasted data against a reference time series based onhistorical data.

To try using a TimesFM model with theAI.FORECAST function, seeForecast a time series with a TimesFM univariate model.

This feature isgenerally available(GA).

BigQuery now supports TransUnion forentity resolution.This feature isgenerally available(GA).

You can now usevisualization cells toautomaticallygenerate a visualizationof any DataFrame in your notebook.You can customize the columns, chart type, aggregations, colors , labels, andtitle.

This feature is inPreview.

In BigQuery ML, you can now fully manage open models as Vertex AI endpoints.BigQuery-managed open models offer the following benefits:

• Manage Vertex AI resource by using SQL queries
• Automatic or immediate open model undeploymentto save costs
• Customize model deployment machine typesor reserve open model resources byusing Compute Engine reservations

This feature is inPreview.

The following features are nowgenerally available (GA) in BigQuery Studio:

• To streamline resource discovery and access, theleft Explorerpane has been reorganizedinto three sections: Explorer, Classic Explorer, and Git repository. You canstill use the Classic Explorer, which provides the complete resources tree.

• In the Explorer pane, you can use thesearchfeature to find BigQueryresources in your organization. The results appear in a new tab in thedetails pane. You can use filters to narrow your search.

• You can access job histories by clickingJobhistory in the Explorer pane. A new tab opensthat displays a list of job histories. BigQuery Studio no longer has abottom pane for job history.

• To reduce tab proliferation, clicking a resource opens it within the sametab. To open the resource in a separate tab, pressCtrl (orCommand on macOS) and click the resource. To prevent the currenttab from getting its content replaced, double-click the tab. The namechanges from italicized to regular font. If you stilllose your resource, you can clicktab_recent Recent tabs in thedetails pane to find the resource.

• You can use breadcrumbs to navigate through different tabs andresources in the details pane.

• In the Home tab, theWhat's new sectioncontains a list of new capabilities and changes to the BigQuery Studio.

• The notebook action bar is consolidated by default to give you more screenspace for writing code.

You can now access repositories by clicking Repositories in the Explorer pane.A new tab opens that displays a list of repositories. The Explorer pane nolonger has a bottom pane for repositories. When you open a workspace in arepository, it opens in the Git repository pane in the left pane. Thesefeatures are available in BigQuery Studio inpreview.

You canvisualize your geospatial query resultson an interactive map in BigQuery Studio. This feature isgenerally available(GA).

You can use thedbt-bigquery adapter to run Python code that's defined inBigQuery DataFrames. For more information, seeUse BigQuery DataFrames in dbt.This feature isgenerally available(GA).

You can nowembed natural language as comments in existing SQL to refine andtransform your code.This feature ispreview.

The BigQuery Data Transfer API (bigquerydatatransfer.googleapis.com) is nowenabled by default for every new Google Cloud project. This feature isgenerally available(GA).

You can now useSQL cells to write,edit, and run SQL queries on your BigQuery data directly from yournotebooks. This feature is inPreview.

An updated version of theODBC driver for BigQuery is now available.

You canallocate idle slots fairly acrossreservations within a single admin project. This ensures each reservationreceives an approximately equal share of available capacity. This feature is nowgenerally available(GA).

You can set amaximum slotlimit for areservation. You can configure the maximum reservation size when creating orupdating a reservation. This feature is nowgenerallyavailable (GA).

Security, privacy, and compliance for Gemini inBigQuery details howcustomer data is protected and processed by Gemini in BigQuery.

Thedefault limit ofQueryUsagePerDay foron-demand pricing has changed. The default limit of all new projects is now 200TiB. For existing projects, the default limit has been set based on yourproject's usage over the last 30 days. Projects that havecustom costcontrols configured or that usereservations aren't affected.If the new limit might affect your workload, create acustom costcontrol based on your workload needs.

You can setlabels on reservations. These labelscan be used to organize your reservations and for billing analysis. This featureisgenerallyavailable (GA).

You canspecify which reservation a query uses atruntime, andset IAMpolicies directly onreservations. Thisprovides more flexibility and fine-grained control over resource management.This feature isgenerallyavailable (GA).

As of February 25, 2025, enhancements to theworkload managementautoscaler that were announced onJuly31, 2024 have rolled out to all users.These enhancements aregenerallyavailable (GA).

The BigQuery Data Transfer Service can now transfer data from thefollowing data sources:

• PayPal
• Stripe

Transfers from these data sources are supported inpreview.

The BigQuery Data Transfer Service can nowtransfer reporting data from Google Analytics 4into BigQuery. You can also include custom reports fromGoogle Analytics 4 in your data transfer. This feature isgenerally available (GA).

You can now set the priority of BigQuery jobs initiated byDataform workflows to run queries as interactive jobs that startrunning as quickly as possible or as batch jobs with lower priority. For moreinformation, seeCreate a pipeline scheduleandInvocationConfig.This feature isgenerally available(GA).

TheINFORMATION_SCHEMA.SHARED_DATASET_USAGE viewnow includes the following schema fields to support usage metrics forexternal tables and routines:

• shared_resource_id: the ID of the queried resource
• shared_resource_type: the type of the queried resource
• referenced_tables: Containsproject_id,dataset_id,table_id, andprocessed_bytesfields of the base table.

These fields aregenerally available (GA).

Starting March 17, 2026, the BigQuery Data Transfer Service will require thebigquery.datasets.setIamPolicy and thebigquery.datasets.getIamPolicypermissions on the target dataset to create or update a transfer configuration.For more information, seeChanges to dataset-level access controls.

You can now use thenotebook galleryin the BigQuery web UI as your central hub for discovering and using prebuilt notebooktemplates. This feature is inpreview.

You can now applySQL query generated in the Gemini Cloud Assist chatto the query open in your editor. This feature is inPreview.

You can now create BigQuerynon-incremental materialized views over Spanner datato improve query performance by periodically caching results.This feature is inPreview.

To simplify access management for your Iceberg tables, youcan usecredential vending mode with the Apache Iceberg REST catalog in BigLake metastore. Credential vending removesthe need for catalog users to have direct access to Cloud Storage buckets. Thisfeature is inPreview.

BigQuery data preparation supports unnesting arrays, which expands eacharray element into its own row for easier analysis. For more information, seeUnnest arrays.This feature isgenerallyavailable (GA).

History-based query optimizationsare now enabled by default. If history-based optimizations have been previouslydisabled, you canre-enable history-based optimizationsfor your project or organization.

TheARRAY_FIRST,ARRAY_LAST,andARRAY_SLICEGoogleSQL functions are nowgenerally available (GA).

BigQuerydatacanvas nowsupports destination table nodes. Destination table nodes let you persist queryresults to a new or existing table. This feature isgenerallyavailable (GA).

For command-line users, BigQuery is now integrated with the Gemini CLI to provide an agentic CLI experience. Using the dedicatedGemini CLI extensions for BigQuery, you can search, explore, analyze, and gain insights from your data by asking natural language questions, generating forecasts, and running contribution analysis directly from the command line. This feature is available in beta.

BigQuery ML now supportsvisualization of model monitoring metrics.This feature lets you use charts and graphs toanalyze model monitoring function output.You can use metric visualization with theML.VALIDATE_DATA_SKEWandML.VALIDATE_DATA_DRIFTfunctions. This feature isgenerally available(GA).

You can now run federated queries againstPostgreSQL dialect databases in Spanner usingBigQuery external datasets with GoogleSQL; this includescross-region federated queries. This feature isgenerally available (GA).

SQL code completion is now available for all BigQuery projects. To learn how to enable and activate Gemini in BigQuery features, seeSet up Gemini in BigQuery. This feature is available inpreview.

You can use theJSON_KEYS function to extract unique JSON keys from a JSON expression, and you can specify amode for some JSON functions that take a JSONPath to allow more flexibility in how the path matches the JSON structure. These features aregenerally available (GA).

You can nowaccess snapshots of Apache Iceberg external tables that are retained in your Iceberg metadata by using theFOR SYSTEM_TIME AS OF clause. This feature isgenerally available (GA).

You can now enable theBigQuery advanced runtime to improve query execution time and slot usage. This feature isgenerally available (GA).Between September 15, 2025 and early 2026, the BigQuery advanced runtime will become the default runtime for all projects.

In the BigQuery Studio, in the Explorer pane, you can nowopen saved queries in Connected Sheets. This feature isgenerally available (GA).

Use theBigQuery migration assessment for Informatica to assess the complexity of migrating data from your Informatica platform to BigQuery. This feature is inPreview.

When you use theData Science Agent in BigQuery, you can now use the Apache Spark or PySpark keywords in your prompt. The Data Science Agent is inPreview.

Gemini now recommends natural language prompts for you in theSQL generation tool. This feature is inPreview.

You canload files from Cloud Storage in BigQuery data preparations. This feature is inPreview.

You can now performsupervised tuning on a BigQuery MLremote model based on a Vertex AIgemini-2.5-pro orgemini-2.5-flash-lite model.

Thebatch andinteractive translators now caches your metadata, which can improve latency when you run a SQL translation. This feature isgenerally available (GA).

You can configure reusable, default Cloud resource connections in a project.Default connections aregenerally available (GA).

You can now includetable parameters when you create a table-valued function (TVF). This feature is inPreview.

You can now add tables and views as tasks to BigQuery pipelines. For more information, seeAdd a pipeline task. This feature is inPreview.

When you use theData Science Agent in BigQuery, you can now use the@ symbol to search for BigQuery tables in your project, and you can use the+ symbol to search for files to upload. The Data Science Agent is inPreview.

You canflatten records inBigQuery data preparation with a single operation. This feature isgenerally available (GA).

BigQuery now supportssoft failover with managed disaster recovery. This feature isgenerally available (GA).

TheINFORMATION_SCHEMA.RESERVATIONS_TIMELINE view now includes theper_second_details schema field. This new field provides information regarding reservation capacity and usage on a per-second basis, and also includes details on autoscale utilization. This feature isgenerally available (GA).

You can now create aremote model based on an open embedding model from Vertex Model Garden or Hugging Face that is deployed to Vertex AI. Options include E5 Embedding and other leading open embedding generation models. You can then use theML.GENERATE_EMBEDDING function with this remote model to generate embeddings.

Try this feature with theGenerate text embeddings by using an open model and theML.GENERATE_EMBEDDING function tutorial.

This feature is inPreview.

You can now enable the automatic selection of a processing location in your pipeline configurations. For more information, seeCreate pipelines. This feature isgenerally available (GA).

You can now create aremote model based on the Vertex AIgemini-embedding-001 model. You can then use theML.GENERATE_EMBEDDING function with this remote model to generate embeddings. This feature is inPreview.

You can now reference BigQuery ML and DataFrames in yourprompts when you use theData Science Agent in a BigQuery notebook. The Data Science Agent is inPreview.

You can now configure listings for multiple regions for shared datasets and linked dataset replicas in BigQuery sharing. For more information, seeCreate a listing. This feature is inpreview.

For additional layers of security and control, you can now use query templates to predefine and limit the queries that can be run in data clean rooms. For more information, seeUse query templates. This feature is inpreview.

You candeduplicate table data with Gemini assistance in yourBigQuery data preparations. Deduplication is inPreview.

You can now use data insights to have Geminigenerate table and column descriptions from table metadata. This feature isgenerally available (GA).

You can use theST_REGIONSTATS geography function to combine raster data using Earth Engine with your vector data stored in BigQuery. For more information, seeWork with raster data and try the tutorial that shows you how touse raster data to analyze global temperature. This feature isgenerally available.

Multi-statement transactions are now available forBigLake Iceberg tables in BigQuery. This feature is inPreview.

Starting September 25, 2025, the BigQuery Data Transfer Service for third-party SAAS and database connectors will update to a consumption-based pricing model. With this new pricing model, you will be charged based on the compute resources consumed by your data transfers, measured in slot-hours. For more information, seeData Transfer Service pricing. This pricing update applies to the following third-party connectors when they aregenerally available (GA):

• Facebook Ads
• MySQL
• Oracle
• PostgreSQL
• Salesforce
• Salesforce Marketing Cloud
• ServiceNow
• Other third-party connectors planned for future releases

When you use the Data Science Agent in BigQuery, you can now use thetable selector to choose one or more BigQuery tables to analyze. The Data Science Agent is inPreview.

In the BigQuery console, you can now use theReference panel to do the following:

• In the query editor, you can use theReference panel to preview the schema details of tables, snapshots, views, and materialized views, or open these resources in a new tab. You can also use the panel to construct new queries or edit existing queries by inserting query snippets or field names.

• In the notebook editor, you can use theReference panel to preview the schema details of tables, snapshots, views, or materialized views, or open these resources in a new tab.

This feature isgenerally available (GA).

You can nowvisualize your geospatial query results on an interactive map in BigQuery studio. This feature is inpreview.

You can usecross region federated queries to query Spanner tables from regions other than the source BigQuery region. These cross region queries incur additionalSpanner network egress charges. This feature isgenerally available (GA).

You can aggregate table data with Gemini assistance in yourBigQuery data preparations. Aggregations in data preparations are inPreview.

You can nowsave query results to Cloud Storage. This feature isgenerally available (GA).

BigQueryresource utilization charts aregenerally available (GA).

You can now useWITH expressions in your GoogleSQL queries to create temporary variables. This feature isgenerally available (GA).

You can now usechained function call syntax in GoogleSQL to make deeply nested function calls easier to read. This feature isgenerally available (GA).

BigQuery data preparations are now represented in the SQLX format and in the pipe query syntax to simplify the CI/CD code review process. For more information, seeManage data preparations.

Enabling the advanced runtime now includesshort query optimizations. This feature is inpreview.

You can now use the newData Science Agent (DSA) for Colab Enterprise and BigQuery to automate exploratory data analysis, performmachine learning tasks, and deliver insights all within aColab Enterprise notebook. This feature is inpreview.

You can manage data profile scans and data quality scans across your project by using theMetadata curation page in the Google Cloud console. For more information, seeProfile your data andScan for data quality issues. This feature isgenerally available (GA).

You can now usecontinuous queries toexport BigQuery data to Spanner in real time. This feature is inPreview.

BigQuery ML has improved throughput by more than 100x for the followinggenerative AI functions:

• ML.GENERATE_TEXT
• AI.GENERATE_TABLE
• AI.GENERATE
• AI.GENERATE_BOOL
• AI.GENERATE_DOUBLE
• AI.GENERATE_INT

Actual performance varies based on the number of input and output tokens in therequest, but a typical 6-hour job can now process millions of rows. For moreinformation, seeGenerative AI functions.

BigQuery ML now can automatically detect model quota increases in Vertex AI, and automatically adjusts the quota for any BigQuery ML functions that use those models. You no longer need to email the BigQuery ML team to increase model quota.

The Gemini for Google Cloud API (cloudaicompanion.googleapis.com) is now enabled by default for most BigQuery projects. Exceptions include projects where customers have opted out, and those linked to accounts based in EMEA regions includingBigQuery Europe, Middle East, and Africa regions.

You can now associatedata policies directly on columns. This feature enables direct database administration for controlling access and applying masking and transformation rules at the column level. This feature is inPreview.

You can now use theVECTOR_INDEX.STATISTICS function to calculate how much an indexed table's data has drifted between when avector index was created and the present. If table data has changed enoughto require avector index rebuild, you can use theALTER VECTOR INDEX REBUILD statementto rebuild the vector index. This feature is inPreview.

Access Transparency supportsBigQuery data preparation in theGA stage.

TheCREATE EXTERNAL TABLE andLOAD DATA statements now support the following options inPreview:

• null_markers: define the strings that representNULL values in CSV files.
• source_column_match: specify how loaded columns are matched to the schema. You can match columns by position or by name.

You can now use theMATCH_RECOGNIZE clause in your SQL queries to filter and aggregate matches across rows in a table. This feature is inPreview.

You can now use theDISTINCT pipe operator to select distinct rows from a table in your pipe syntax queries. This feature isgenerally available (GA).

You can now usenamed windows in your pipe syntax queries. This feature isgenerally available (GA).

You can now use theWITH pipe operator to define common table expressions in your pipe syntax queries. This feature isgenerally available (GA).

You can now add comments tonotebooks,data canvases,data preparation files, orsaved queries. You can also reply to existing comments or get a link to them. This feature is inPreview.

You can now createBigQuery ML models by using theGoogle Cloud console user interface. This feature is inPreview.

You can nowcommercialize your BigQuery sharing listings on Google Cloud Marketplace. This feature isgenerally available (GA).

You can flattenJSON columns in BigQuery data preparation with a single operation. This feature isgenerally available (GA).

Starting August 1, 2025, GoogleSQL will become the default dialect for queries run from the command line interface (CLI) or API. To use LegacySQL, you will need to explicitly specify it in your requests orset the configuration settingdefault_sql_dialect_option to'default_legacy_sql' at the project or organization level.

You can now use your Google Account user credentials to authorize the execution of a data preparation in development. For more information, seeManually run a data preparation in development. This feature is inpreview.

BigQuery now supports the following Apache Hadoop migration features inPreview:

• Use thedwh-migration-dumper tool to migrate the metadata necessary for a Hadoop permissions and data migration.
• Migrate permissions from Apache Hadoop, Apache Hive, and Ranger HDFS to BigQuery.
• Migrate tables from a HDFS data lake to Google Cloud.

You can use the@@location system variable to set the location in which to run a query. This feature isgenerally available (GA).

You can nowupdate a Cloud KMS encryption key by updating the table with the same key. This feature isgenerally available (GA).

You can nowcreate and manage scheduled notebooks using theSchedule details pane in BigQuery Studio. This feature isgenerally available (GA).

You can now use theuse thePARTITION BY clause of theCREATE VECTOR INDEX statement topartition TreeAH vector indexes. Partitioning enables partition pruning and can decrease I/O costs. This feature is inpreview.

BigQuery search indexes provide free index management until your organization reaches thelimit in a given region. You can now use theINFORMATION_SCHEMA.SEARCH_INDEXES_BY_ORGANIZATION view to understand your current consumption towards that limit, broken down by projects and tables. This feature isgenerally available (GA).

Colab Enterprise notebooks in BigQuery let you do the following inPreview:

• Explain code with Gemini assistance
• Fix and explain errors with Gemini assistance

You can now use theApache Iceberg REST catalog in BigLake metastore to create interoperability between your query engines by allowing your open source engines to access Iceberg data in Cloud Storage. This feature is inPreview.

You can now use data insights to have Geminigenerate table and column descriptions from table metadata. This feature is inPreview.

You can nowpublish the results of a data quality scan as Dataplex Universal Catalog metadata. Previously, data quality scan results were published only to the Google Cloud console. The latest results are saved to the entry that represents the source table. You can view the results in the Google Cloud console. If you want to enable catalog publishing for an existing data quality scan, you must edit the scan and re-enable the publishing option. This feature isgenerally available (GA).

BigQuery now offers the following Gemini-enhanced SQL translation features:

• CreateGemini-based configuration YAML files to generate AI suggestions for batch or interactive SQL translations. This feature is nowgenerally available (GA).
• After making a batch SQL translation, review your translation output, including Gemini-based suggestions, using thecode tab andconfiguration tab. This feature is nowgenerally available (GA).
• When making an interactive SQL translation,create and apply Gemini-enhanced translation rules to customize your SQL inputs. This feature is inPreview.

In BigQuery ML, you can now forecast multiple time series at once by using theTIME_SERIES_ID_COL option that is available inARIMA_PLUS_XREG multivariate time series models. Try this feature with theForecast multiple time series with a multivariate model tutorial. This feature isgenerally available (GA).

TheMerchant Center best sellers report supports multi-client accounts (MCAs). If you have an MCA, you can use theaggregator_id to query the tables. TheBestSellersEntityProductMapping table maps the best-selling entities to the products in the sub-accounts' inventory. This provides a consolidated view of best-selling products, which you can then join with product data for more detailed insights. This feature isgenerally available (GA).

You can now manageIAM tags on BigQuery datasets and tables using SQL. This feature isgenerally available (GA).

The BigQuery migration assessment is now available for workflows that useCloudera and Apache Hadoop. This feature is inPreview.

Dark theme is now available for BigQuery inPreview. To enable the dark theme, in the Google Cloud console, clickSettings and utilities > Preferences. In the navigation menu, clickAppearance, and then select your color theme and clickSave.

The following GoogleSQL functions are now available inpreview:

• TheARRAY_FIRST function returns the first element of the input array.
• TheARRAY_LAST function returns the last element of the input array.
• TheARRAY_SLICE function returns an array that contains consecutive elements from the input array.

Forsupported Gemini models, you can now useVertex AI Provisioned Throughput with theML.GENERATE_TEXTandAI.GENERATE functions to provide consistent high throughput for requests.

This feature isgenerally available (GA).

An updated version of theODBC driver for BigQuery is now available.

You can referenceIceberg external tables in materialized views instead of migrating that data to BigQuery-managed storage. This feature isgenerally available (GA).

The organization-levelconfiguration settings fordefault_sql_dialect_option andquery_runtime are unsupported.

BigQuery tables for Apache Iceberg have been renamedBigLake tables for Apache Iceberg in BigQuery. This feature is nowgenerally available (GA).

You can now use theBigQuery advanced runtime to improve query execution time and slot usage. This feature is inPreview.

BigQuery metastore has been renamedBigLake metastore and is nowgenerally available (GA). The feature formerly known as BigLake metastore has been renamed BigLake metastore (classic).

In the BigQuery console, in theWelcome tab, you can now try theApache Spark demo notebook that walks you through the basics of Spark notebook and showcasesserverless Spark in BigQuery. This feature isgenerally available (GA).

In the navigation menu, you can now go toSettings and selectConfiguration settings tocustomize the BigQuery Studio experience for users within the selected project or organization. This is achieved by showing or hiding user interface elements. This feature is inpreview.

TheCREATE EXTERNAL TABLE andLOAD DATA statements now support the following options inpreview:

• time_zone: specify a time zone to use when loading data
• date_format,datetime_format,time_format, andtimestamp_format: define how date and time values are formatted in your source files

BigQuery now supports usingSpanner external datasets withauthorized views,authorized routines, andCloud resource connections. This feature isgenerally available (GA).

You can now createevent-driven transfers when transferring data from Cloud Storage to BigQuery. Event-driven transfers can automatically trigger transfer runs when data in your Cloud Storage bucket has been modified or added. This feature isgenerally available (GA).

You can now use your Google Account user credentials to authorize the creation, scheduling, and running of pipelines as well as the scheduling of notebooks and data preparations. For more information, seeCreate a pipeline schedule. This feature is inpreview.

You can now use thedbt-bigquery adapter to run Python code that's defined in BigQuery DataFrames. For more information, seeUse BigQuery DataFrames in dbt. This feature is inpreview.

Column metadata indexing is now available for bothBigQuery tables andexternal tables. This feature isgenerally available (GA).

You can now create aserverless Spark session and run PySpark code in a BigQuery notebook. This feature isgenerally available (GA).

You can nowshare Pub/Sub streaming data through BigQuery sharing with additionalclient libraries support and provider usage metrics. This feature isgenerally available (GA).

BigQuery offersoptional job creation mode to speed up small queries that you use in your dashboards, data exploration, and other workflows. This mode automatically optimizes eligible queries and uses a cache to improve latency. This feature isgenerally available (GA).

When youSet up Gemini in BigQuery you are now prompted to grant the BigQuery Studio User and BigQuery Studio Admin roles. These roles now include permission to use Gemini in BigQuery features. This feature isgenerally available (GA).

You can usecustom constraints with Organization Policy to provide more granular control over specific fields for some BigQuery resources. This feature is available inPreview.

When you migrate Teradata data to BigQuery using the BigQuery Data Transfer Service, you can nowspecify the outputs of the BigQuery translation engine to use as schema mapping. This feature is inpreview.

Starting March 17 2026, thebigquery.datasets.getIamPolicyIAM permission is required to view a dataset's access controls and to query theINFORMATION_SCHEMA.OBJECT_PRIVILEGESview. Thebigquery.datasets.setIamPolicy permission is required to update adataset's access controls or tocreate a dataset with access controls using theAPI. For more information on this change and how to opt into early enforcement, seeChanges to dataset-level access controls.

You can select multiple columns and perform data preparation tasks on them, including dropping columns. For more information, seePrepare data with Gemini. This feature isgenerally available (GA).

You are now able toset access controls on routines. This feature is inPreview.

You can now performsupervised tuning on a BigQuery MLremote model based on a Vertex AIgemini-2.0-flash-001 orgemini-2.0-flash-lite-001 model.

Spanner now supportscross regional federated queries from BigQuery which allow BigQuery users to query Spanner tables from regions other than their BigQuery region. Users don't incurSpanner network egress charges during the preview period. This feature is inpreview.

Continuous queries let you build long-lived, continuously processing SQL statements that can analyze, process, and perform machine learning (ML) inference on incoming data in BigQuery in real time.

• To monitor your continuous queries, you can use acustom job ID prefix to simplify filtering or viewmetrics specific to continuous queries in Cloud Monitoring.
• Continuous queries can useslot autoscaling to dynamically scale allocated capacity to accommodate your workload.

This feature isgenerally available (GA).

BigQuery now supports cross-region transfers forbatch loading andexporting data. You can load or export your data from any region or multi-region to any other region or multi-region using a singlebq load,LOAD DATA,bq extract, orEXPORT DATA statement. This feature isgenerally available (GA).

Vector indexes support theTreeAH index type, which uses Google's ScaNN algorithm. The TreeAH index is optimized for efficient batch processing, capable of handling anywhere from a few thousand to hundreds of thousands of embeddings at once. This feature isgenerally available (GA).

You can nowschedule automated data transfers from Snowflake to BigQuery using the BigQuery Data Transfer Service. This feature is inpreview.

The following SQL features are nowgenerally available (GA) in BigQuery:

• GROUP BY STRUCT and theSELECT DISTINCT clause.
• GROUP BY ARRAY and theSELECT DISTINCT clause.
• GROUP BY ALL clause.

You can now use BigQuery and BigQuery DataFrames to enable multimodal analysis, transformation, and data engineering (ELT) workflows in both SQL and Python. Use multimodal data features to do the following:

• Integrate unstructured data into standard tables by usingObjectRef values, and then work with this data in analysis and transformation workflows by usingObjectRefRuntime values.

• Use generative AI to analyze multimodal data and generate embeddings by usingBigQuery ML SQL functions orBigQuery DataFrames methodswith Gemini and multimodal embedding models.

• Create multimodal DataFrames in BigQuery DataFrames, and then useobject transformation methods to transform images and chunk PDF files.

• Use Python user-defined functions (UDFs) to transform images and chunk PDF files.

This feature is inPreview.

BigQuery resource utilization charts have the following changes:

• The default timeline shown in the event timeline chart has changed from one to six hours.
• Several improvements have been made to the views, including a new reservation slot usage view. This view helps monitor idle, baseline, and autoscaled slot usage.

This feature is inPreview.

You can now view theQuery text section in a BigQuery execution graph to understand how the stage steps are related to the query text. This feature is inpreview.

In the Google Cloud console, Analytics Hub has been renamedBigQuery sharing (Analytics Hub).

Changes that you make to your saved queries are nowautomatically saved. This feature is inpreview.

When you translate SQL queries from your source database, you can use configuration YAML files tooptimize and improve the performance of your translated SQL. This feature isgenerally available (GA).

Dataplex automatic discovery in BigQuery scans your data in Cloud Storage buckets to extract and catalog metadata, creating BigLake, external, or object tables for analytics and AI for insights, security, and governance. This feature isgenerally available (GA).

You can now work with aGemini powered assistant in a BigQuery data canvas. The data canvas assistant is an agent-like tool, capable of constructing and modifying a data canvas to answer data analytics questions from user prompting. This feature is now inPreview.

You can nowallocate idle slots fairly across reservations within a single admin project. This ensures each reservation receives an approximately equal share of available capacity. This feature is inpublic preview.

You can now set amaximum slot limit for a reservation. You can configure the maximum reservation size when creating or updating a reservation. This feature is inpublic preview.

You can nowspecify which reservation a query uses at runtime, andset IAM policies directly on reservations. This provides more flexibility and fine-grained control over resource management. This feature is inpublic preview.

You can get the required permissions to use BigQuery data preparation through theBigQuery Studio User (roles/bigquery.studioUser) andGemini for Google Cloud User (roles/cloudaicompanion.user) roles, and permission to access the data you're preparing.

BigQuery data preparation no longer requires that you have the permissions granted by the following IAM roles:

• BigQuery Data Editor(roles/bigquery.dataEditor)
• Service Usage Consumer (roles/serviceusage.serviceUsageConsumer)

For more information about the required roles, seeManage data preparations.

You can now enablefine-grained access control on BigQuery metastore Iceberg tables. This feature isgenerally available (GA).

BigQuery now provides spend-basedcommitted use discounts (CUDs). Spend-based committed use discounts provide a discount in exchange for your commitment to spend a minimum amount per hour on PAYG compute resourceslisted here. You can purchase CUDs with a one or three year commitment period.

You can now useBigQuery DataFrames version 2.0, which makes security and performance improvements to the BigQuery DataFrames API, adds new features, and introduces breaking changes.

You can usepartial ordering mode in BigQuery DataFrames to generate efficient queries. This feature isgenerally available (GA).

You can now combine raster and vector data with theST_REGIONSTATS geography function to perform geospatial analysis in BigQuery. For more information, seeWork with raster data and try the tutorial that shows you how touse raster data to analyze global temperature by country. This feature is inpreview.

Analytics Hub has been renamedBigQuery sharing. You'll see this new name in the documentation set and the marketing collateral. The product functionality and endpoints remain the same. For more information, seeIntroduction to data governance in BigQuery.

You can now use the Apache Arrow format tostream data to BigQuery with the Storage Write API. This feature is available inpreview.

Updated pricing, packaging, and setup guidance is now available forGemini in BigQuery.

Dataplex Catalog has been renamedBigQuery universal catalog. You'll see this new name in the product page of the Google Cloud console, the documentation set, and the marketing collateral. Universal catalog brings together thedata catalog capabilities of Dataplex Catalog and theruntime metastore capabilities of BigQuery metastore. For more information, seeIntroduction to data governance in BigQuery.

BigQuery ML now offers a built-inTimesFM univariate time series forecasting model that implements Google Research's open source TimesFM model. You can use BigQuery ML's built-in TimesFM model with theAI.FORECAST function to perform forecasting without having to create and train your own model. This lets you avoid the need for model management.

To try using a TimesFM model with theAI.FORECAST function, seeForecast a time series with a TimesFM univariate model.

This feature is inpreview.

You can nowconnect BigQuery metastore to Apache Flink. This feature isgenerally available (GA).

You can nowcreate, view, modify, and delete Apache Iceberg resources in BigQuery metastore. This feature isgenerally available (GA).

BigQuery ML now uses dynamic token-based batching forembedding generation requests. Dynamic token-based batching puts as many rows as possible into one request. This change boosts per-request utilization and improves scalability for anyqueries per minute (QPM) quota. Actual performance varies based on the embedding content length, with an average 10x improvement.

Smart-tuning is now supported formaterialized views when they are in the same project as one of their base tables, or when they are in the project running the query. This feature isgenerally available (GA).

BigQuery data preparation isgenerally available (GA). It offers AI-powered suggestions from Gemini for data cleansing, transformation, and enrichment. BigQuery supports visual data preparation pipelines and pipeline scheduling with Dataform.

An updated version ofJDBC driver for BigQuery is now available.

You can now createremote models in BigQuery ML based onLlama andMistral AI models in Vertex AI.

Use theML.GENERATE_TEXT function with these remote models to perform generative natural language tasks for text stored in BigQuery tables. Try this feature with theGenerate text by using theML.GENERATE_TEXT function tutorial.

This feature isgenerally available (GA).

BigQuery ML now supports the followinggenerative AI functions, which let you analyze text using a Vertex AI Gemini model. The function output includes a response that matches the type in the function name:

• AI.GENERATE
• AI.GENERATE_BOOL
• AI.GENERATE_INT
• AI.GENERATE_DOUBLE

This feature is inpreview.

You can now generate structured data by using BigQuery ML'sAI.GENERATE_TABLE function with Gemini 1.5 Pro, Gemini 1.5 Flash, and Gemini 2.0 Flash models. You can use theAI.GENERATE_TABLE function'soutput_schema argument to more easily format the model's response. Theoutput_schema argument lets you specify a SQL schema for formatting, similar to the schema used in theCREATE TABLE statement. By creating structured output, you can more easily convert the function output into a BigQuery table.

Try this feature with theGenerate structured data by using theAI.GENERATE_TABLE function tutorial.

This feature is inpreview.

BigQuery migration assessment now includes support for Amazon Redshift Serverless. This feature is in preview.

You can nowgenerate Dataframes code in BigQuery Notebooks that use BigFrames libraries. In your code generation prompt, include the wordBigFrames to generate code that usesBigQuery DataFrames. This feature is inpreview.

The Python code that you generate usingGemini in BigQuery Notebooks is now much more likely to leverage your data. With this change, BigQuery Notebooks can intelligently pull relevant table names directly from your BigQuery project, resulting in personalized, executable Python code.

You can now create and usePython user-defined functions (UDFs) in BigQuery. Python UDFs support the use of additional libraries and external APIs. This feature is inpreview.

Pipe syntax supports a linear query structure designed to make your queries easier to read, write, and maintain. This feature isgenerally available (GA).

You can use aCREATE MODEL statement to create acontribution analysis model in BigQuery ML. Thetop_k_insights_by_apriori_support andpruning_method model options are now supported. You can use a contribution analysis model with theML.GET_INSIGHTS function to generate insights about changes to key metrics in your multi-dimensional data. The following metric types are supported:

• Summable metric
• Summable ratio metric
• Summable by category metric

This feature isgenerally available (GA).

Iceberg external tables now support merge-on-read. You can query Iceberg tables with position deletes and equality deletes. This feature isgenerally available (GA).

You can now define a_CHANGE_SEQUENCE_NUMBER for BigQuery change data capture (CDC) to manage streamingUPSERT ordering for BigQuery. This feature isgenerally available (GA).

On the Scheduling page, you can now view existing schedules, create new schedules, and perform other actions for data preparations, notebooks, BigQuery pipelines, and scheduled queries. For more information, seeCreate a pipeline schedule. This feature isgenerally available (GA).

You can now useBigQuery Data Transfer Service for Search Ads to viewPerformance Max (PMax) campaign data for the following tables:

• CartDataSalesStats
• ProductAdvertised
• ProductAdvertisedDeviceStats
• ProductAdvertisedConversionActionAndDeviceStats

This feature isgenerally available (GA).

You can now skip loading match tables forBigQuery Data Transfer Service for Google Ad Manager. If match tables are not needed, you can set parameterload_match_tables toFALSE. This feature isgenerally available (GA).

You canbuild BigQuery pipelines (formerly workflows), composed of SQL queries or notebooks, in BigQuery Studio. You can then run these pipelines on a schedule. You can also configure notebook runtimes for a pipeline, share a pipeline, or share a pipeline link. This feature isgenerally available (GA).

You can now configure the repeat frequency ofBigQuery Data Transfer Service for Google Ad Manager. This option has a default of every 8 hours and a minimum of every 4 hours. This feature isgenerally available (GA).

BigQuery now supportssubqueries inrow level access policies. It also includes support forBigLake managed tables and theBigQuery Storage Read API. This feature is nowgenerally available (GA).

You can include data preparation tasks inBigQuery pipelines that execute your code assets in sequence at a scheduled time. This feature is inPreview.

You can nowenable metadata caching for SQL translation, which can significantly reduce latency for subsequent translation requests. This feature is inpreview.

You can now set thecolumn granularity when youcreate a search index, which stores additional column information in your search index to further optimize your search query performance. This feature is inpreview.

BigQuery ML now supportsvisualization of model monitoring metrics. This feature lets you use charts and graphs toanalyze model monitoring function output. The following functions support metric visualization:

• ML.VALIDATE_DATA_SKEW: compute the statistics for a set of serving data, and then compare them to the statistics for the data used to train a BigQuery ML model in order to identify anomalous differences between the two data sets.
• ML.VALIDATE_DATA_DRIFT: compute and compare the statistics for two sets of serving data in order to identify anomalous differences between the two data sets.

This feature is inpreview.

You can now useKLL quantile functions to efficiently compute approximate quantiles. This feature is inpreview.

The BigQuery Data Transfer Service can nowtransfer reporting and configuration data from Google Analytics 4 into BigQuery. This feature is inpreview.

You can now setlabels on reservations. These labels can be used to organize your reservations and for billing analysis. This feature is inpreview.

We have redesigned theAdd Data dialog to guide you through loading data into BigQuery with a source-first experience and enhanced search and filtering capabilities. This feature isgenerally available (GA).

BigQuery workflows have been renamed to BigQuery pipelines in the Google Cloud console. For more information, seeIntroduction to BigQuery pipelines.

You can now userepositories andworkspaces in BigQuery to perform version control.

Repositories perform version control on files by using Git to record changes and manage file versions. You can use workspaces within repositories to edit the code stored in the repository.

You can have a repository use Git directly on BigQuery, or you canconnect a repository to a third-party Git provider.

This feature is inpreview.

You can now createremote models in BigQuery ML based on theAnthropic Claude model in Vertex AI.

Use theML.GENERATE_TEXT function with these remote models to perform generative natural language tasks for text stored in BigQuery tables. Try this feature with theGenerate text by using theML.GENERATE_TEXT function tutorial.

You can also evaluate Claude models by using theML.EVALUATE function.

This feature isgenerally available (GA).

You can now create anexternal dataset in BigQuery that links to an existing database inSpanner. This feature isgenerally available (GA).

You can now use theTYPEOF function to determine the data type of an expression. This feature isgenerally available (GA).

You can now useEXPORT DATA statements toreverse ETL BigQuery data to Spanner. This feature isgenerally available (GA).

You can now use Gemini Cloud Assist chat to generateSQL queries andPython code. This feature is inpreview.

Dataform now supports theCMEK organization policy.

You can configure reusable, default Cloud resource connections in a project.Default connections are available inPreview.

An updated version ofODBC driver for BigQuery is now available.

Analytics Hubegress controls anddata clean room subscriptions are now available in all BigQueryeditions and on-demand pricing.

BigQuery Data Transfer Service now supportscustom reports for Google Ads. You can use Google Ads Query Language (GAQL) queries in your transfer configuration to ingest custom Google Ads reports and fields beyond those available in thestandard reports and fields. This feature is nowgenerally available (GA).

BigQuery is now available in theStockholm (europe-north2) region.

You can create aSQL user-defined aggregate function by using theCREATE AGGREGATE FUNCTION statement. This feature isgenerally available (GA).

Gemini in BigQuery can help youcomplete Python code with contextually appropriate recommendations that are based on content in the query editor. This feature is nowgenerally available (GA).

You can now see alist of BigQuery API and service dependencies. You can also review the effects of disabling an API or service.

You can use thebest sellers andprice competitiveness migration guides to transition to the newer version of the reports. This feature is inpreview.

BigQueryresource utilization charts provide metrics views and more chart configuration options inPreview.

You can now use the@@location system variable to set the location in which to run a query. This feature is inpreview.

Subscriber email logging lets you log the principal identifiers of users who execute jobs and queries against linked datasets. You can enable logging at thelisting level and thedata exchange level (for all the listings in the data exchange). Once you enable and save subscriber email logging, this setting cannot be edited. This feature is inpreview.

BigQuery data preparation providescontext-aware join operation recommendations from Gemini. Data preparation is available inPreview.

You can create aJavaScript user-defined aggregate function by using theCREATE AGGREGATE FUNCTION statement. This feature isgenerally available (GA).

The BigQuery MLML.BUCKETIZE andML.QUANTILE_BUCKETIZE functions now support formatting of the function output. You can use theoutput_format argument to format the function output as one of the following:

• A string in the formatbin_<bucket_index>
• A string ininterval notation
• A JSON-formatted string

You can now use theBY NAME andCORRESPONDING modifiers with set operations to match columns by name instead of by position. This feature isgenerally available (GA).

You can now viewstored column usage information for a query job that performs vector search using stored columns. This feature isgenerally available (GA).

We previously communicated that after January 27, 2025, a purchase would be required to useGemini in BigQuery features. We are temporarily delaying enforcement of these procurement methods, and no purchase is required at this time. For more information, seeGemini for Google Cloud pricing.

The following BigQuery ML generative AI features are now available:

• Creating aremote modelbased on anopen model from Vertex Model Garden or Hugging Face that is deployed to Vertex AI.Options include Llama, Gemma, and other leading open text generation models.
• Using theML.GENERATE_TEXT functionwith this remote model to perform a broad range of generative AI tasks.
• Using theML.EVALUATE functionto evaluate the remote model.

Try these features with theGenerate text by using theML.GENERATE_TEXT functionhow-to topic and the
Generate text by using a Gemma open model and theML.GENERATE_TEXT functiontutorial.

These features aregenerally available(GA).

You can now setconditional IAM access on BigQuery datasets with access control lists (ACLs). This feature isgenerally available (GA).

BigQuery metastore lets you access and manage metadata from a variety of processing engines, including BigQuery and Apache Spark. BigQuery metastore supports BigQuery tables and open formats such as Apache Iceberg. This feature is inpreview.

Data preparation in BigQuery lets you test data preparations you're developing before you deploy and schedule runs in production. For more information, seeDevelop a data preparation.

In BigQuery ML, you can now evaluate Anthropic Claude models by using theML.EVALUATE function.The quotasfor use of Anthropic Claude models in BigQuery ML have also been brought intoparity with Vertex AI quotas.

This feature is inpreview.

You can use natural language toprepare data with Gemini in BigQuery.

The BigQuery Data Transfer Service can now transfer data from the following data sources:

• MySQL
• PostgreSQL

Transfers from these data sources are supported inPreview.

In thenavigation menu, you can now go to theSettings page to set default settings that are applied when you start a session in BigQuery Studio. This feature is inpreview.

We have rearranged thenavigation menu into new categories. This feature isgenerally available (GA).

TheBigQuery migration assessment for Oracle now includes a total cost of ownership (TCO) calculator that provides an estimation of compute and storage costs for migrating your Oracle data warehouse to BigQuery. This feature is inpreview.

In BigQuery ML, you can now forecast multiple time series at once by using thenewTIME_SERIES_ID_COL optionthat is available inARIMA_PLUS_XREG multivariate time series models. Try thisfeature with theForecast multiple time series with a multivariate modeltutorial.

This feature is inpreview.

You can now use BigQuery Omni Virtual Private Cloud (VPC) allowlists to restrict access toAWS S3 buckets andAzure Blob Storage from specific BigQuery Omni VPCs. This feature isgenerally available (GA).

An updated version ofJDBC driver for BigQuery is now available.