BigQuery locations

This page explains the concept of location and the different regions where data can be stored and processed. Pricing for storage and analysis is also defined by location of data and reservations. For more information about pricing for locations, see BigQuery pricing. To learn how to set the location for your dataset, see Create datasets. For information about reservation locations, see Managing reservations in different regions.

For more information about how the BigQuery Data Transfer Service uses location, see Data location and transfers.

Locations and regions

BigQuery provides two types of data and compute locations:

A region is a specific geographic place, such as London.
A multi-region is a large geographic area, such as the United States or Europe, that contains many unique and discrete regions. Multi-region locations can provide larger quotas than single regions, but multi-regions don't provide regional redundancy. Data is stored in a single region and compute is only provided within that region. For cross-region redundancy BigQuery offers managed disaster recovery.

For either location type, BigQuery automatically stores copies of your data in two different zones within a single region in the selected location. Multi-regions are considered separate from other regions, even when located within the same zone. For more information about data availability and durability, see Disaster planning.

Supported locations

BigQuery datasets can be stored in the following regions and multi-regions. For more information about regions and zones, see Geography and regions.

Regions

The following table lists the regions in the Americas where BigQuery is available.

Region description	Region name	Details
Columbus, Ohio	`us-east5`
Dallas	`us-south1`	Low CO₂
Iowa	`us-central1`	Low CO₂
Los Angeles	`us-west2`
Las Vegas	`us-west4`
Mexico	`northamerica-south1`
Montréal	`northamerica-northeast1`	Low CO₂
Northern Virginia	`us-east4`
Oklahoma	`us-central2`	Low CO₂
Oregon	`us-west1`	Low CO₂
Salt Lake City	`us-west3`
São Paulo	`southamerica-east1`	Low CO₂
Santiago	`southamerica-west1`	Low CO₂
South Carolina	`us-east1`
Toronto	`northamerica-northeast2`	Low CO₂

The following table lists the regions in Asia Pacific where BigQuery is available.

Region description	Region name	Details
Bangkok	`asia-southeast3`
Delhi	`asia-south2`
Hong Kong	`asia-east2`
Jakarta	`asia-southeast2`
Melbourne	`australia-southeast2`
Mumbai	`asia-south1`
Osaka	`asia-northeast2`
Seoul	`asia-northeast3`
Singapore	`asia-southeast1`
Sydney	`australia-southeast1`
Taiwan	`asia-east1`
Tokyo	`asia-northeast1`

The following table lists the regions in Europe where BigQuery is available.

Region description	Region name	Details
Belgium	`europe-west1`	Low CO₂
Berlin	`europe-west10`
Finland	`europe-north1`	Low CO₂
Frankfurt	`europe-west3`
London	`europe-west2`	Low CO₂
Madrid	`europe-southwest1`	Low CO₂
Milan	`europe-west8`
Netherlands	`europe-west4`	Low CO₂
Paris	`europe-west9`	Low CO₂
Stockholm	`europe-north2`	Low CO₂
Turin	`europe-west12`
Warsaw	`europe-central2`
Zürich	`europe-west6`	Low CO₂

The following table lists the regions in the Middle East where BigQuery is available.

Region description	Region name	Details
Dammam	`me-central2`
Doha	`me-central1`
Tel Aviv	`me-west1`

The following table lists the regions in Africa where BigQuery is available.

Region description	Region name	Details
Johannesburg	`africa-south1`

Multi-regions

Note: Later this year, BigQuery will stop using the term multi-region. Instead, the US and EU region names will be included in the list of single regions and will colocate with the us-central1 and europe-west4 regions respectively. There is no change to existing data residency.

The following table lists the multi-regions where BigQuery is available. When you select a multi-region, you let BigQuery select a single region within the multi-region where your data is stored and processed.

Multi-region description	Multi-region name
Data centers within member states of the European Union¹	`EU`
Data centers in the United States²	`US`

¹ Data located in the EU multi-region is only stored in one of the following locations: europe-west1 (Belgium) or europe-west4 (Netherlands). The exact location in which the data is stored and processed is determined automatically by BigQuery.

² Data located in the US multi-region is only stored in one of the following locations: us-central1 (Iowa), us-west1 (Oregon), or us-central2 (Oklahoma). The exact location in which the data is stored and processed is determined automatically by BigQuery.

BigQuery Studio code asset locations

BigQuery Studio lets you save, share, and manage versions of code assets such as notebooks and saved queries.

The following table lists the regions where BigQuery Studio is available:

	Region description	Region name	Details
Africa
	Johannesburg	`africa-south1`
Americas
	Columbus	`us-east5`
	Dallas	`us-south1`	Low CO₂
	Iowa	`us-central1`	Low CO₂
	Los Angeles	`us-west2`
	Las Vegas	`us-west4`
	Montréal	`northamerica-northeast1`	Low CO₂
	N. Virginia	`us-east4`
	Oregon	`us-west1`	Low CO₂
	São Paulo	`southamerica-east1`	Low CO₂
	South Carolina	`us-east1`
Asia Pacific
	Hong Kong	`asia-east2`
	Jakarta	`asia-southeast2`
	Mumbai	`asia-south1`
	Seoul	`asia-northeast3`
	Singapore	`asia-southeast1`
	Sydney	`australia-southeast1`
	Taiwan	`asia-east1`
	Tokyo	`asia-northeast1`
Europe
	Belgium	`europe-west1`	Low CO₂
	Finland	`europe-north1`	Low CO₂
	Frankfurt	`europe-west3`
	London	`europe-west2`	Low CO₂
	Madrid	`europe-southwest1`	Low CO₂
	Milan	`europe-west8`
	Netherlands	`europe-west4`	Low CO₂
	Turin	`europe-west12`
	Warsaw	`europe-central2`
	Zürich	`europe-west6`	Low CO₂
Middle East
	Dammam	`me-central2`
	Doha	`me-central1`
	Tel Aviv	`me-west1`

BigQuery Omni locations

BigQuery Omni processes queries in the same location as the dataset that contains the tables you're querying. After you create the dataset, the location cannot be changed. Your data resides within your AWS or Azure account. BigQuery Omni regions support Enterprise edition reservations and on-demand compute (analysis) pricing. For more information about editions, see Introduction to BigQuery editions.

	Region description	Region name	Colocated BigQuery region
AWS
	AWS - US East (N. Virginia)	`aws-us-east-1`	`us-east4`
	AWS - US West (Oregon)	`aws-us-west-2`	`us-west1`
	AWS - Asia Pacific (Seoul)	`aws-ap-northeast-2`	`asia-northeast3`
	AWS - Asia Pacific (Sydney)	`aws-ap-southeast-2`	`australia-southeast1`
	AWS - Europe (Ireland)	`aws-eu-west-1`	`europe-west1`
	AWS - Europe (Frankfurt)	`aws-eu-central-1`	`europe-west3`
Azure
	Azure - East US 2	`azure-eastus2`	`us-east4`

BigQuery ML locations

The following sections describe supported locations for BigQuery ML models.

Locations for remote models

This section contains information about supported locations for remote models, and about where remote model processing occurs.

Regional locations

See the following documentation for supported locations for remote models over Google models and partner models:

For Gemini model and embedding model supported regions, see Google model endpoint locations.
For Claude, Llama, and Mistral AI model supported regions, see Google Cloud partner model endpoint locations.

The following table shows which regions are supported for remote models over Cloud AI services and custom models deployed to Agent Platform. The column name indicates the type of remote model.

	Region description	Region name	Vertex AI deployed models	Speech-to-Text API
Americas
	Columbus, Ohio	`us-east5`
	Dallas	`us-south1`	●
	Iowa	`us-central1`	●	●
	Las Vegas	`us-west4`	●
	Los Angeles	`us-west2`	●
	Mexico	`northamerica-south1`
	Montréal	`northamerica-northeast1`	●
	Northern Virginia	`us-east4`	●
	Oregon	`us-west1`	●	●
	Salt Lake City	`us-west3`	●
	São Paulo	`southamerica-east1`	●
	Santiago	`southamerica-west1`
	South Carolina	`us-east1`	●	●
	Toronto	`northamerica-northeast2`	●
Europe
	Belgium	`europe-west1`	●	●
	Finland	`europe-north1`
	Frankfurt	`europe-west3`	●	●
	London	`europe-west2`	●	●
	Madrid	`europe-southwest1`
	Milan	`europe-west8`	●
	Netherlands	`europe-west4`	●	●
	Paris	`europe-west9`	●
	Stockholm	`europe-north2`
	Turin	`europe-west12`
	Warsaw	`europe-central2`	●
	Zürich	`europe-west6`	●
Asia Pacific
	Bangkok	`asia-southeast3`
	Delhi	`asia-south2`
	Hong Kong	`asia-east2`	●
	Jakarta	`asia-southeast2`	●
	Melbourne	`australia-southeast2`
	Mumbai	`asia-south1`	●	●
	Osaka	`asia-northeast2`
	Seoul	`asia-northeast3`	●
	Singapore	`asia-southeast1`	●	●
	Sydney	`australia-southeast1`	●	●
	Taiwan	`asia-east1`	●
	Tokyo	`asia-northeast1`	●	●
Middle East
	Dammam	`me-central2`
	Doha	`me-central1`
	Tel Aviv	`me-west1`	●

If the dataset in which you are creating the remote model is in a single region, the Agent Platform model endpoint must be in the same region. If you specify the model endpoint URL, use the endpoint in the same region as the dataset. For example, if the dataset is in the us-central1 region, then specify the endpoint https://us-central1-aiplatform.googleapis.com/v1/projects/myproject/locations/us-central1/publishers/google/models/<target_model>. If you specify the model name, BigQuery ML automatically chooses the endpoint in the correct region.

Multi-regional locations

Multi-regional support for remote models is as follows:

Gemini models are supported in the US and EU multi-regions.
Claude, Llama, and Mistral AI models in the US multi-region can use the Agent Platform endpoint for any single region within the US multi-region. Claude, Llama, and Mistral AI models in the EU multi-region can use the Agent Platform endpoint for any single region within the EU multi-region except for eu-west2 and eu-west6.
Vertex AI deployed models aren't supported in either multi-region.
Cloud AI services are supported in the US and EU multi-regions.

If the dataset in which you are creating the remote model is in a multi-region, then the Agent Platform model endpoint must be in a region within that multi-region. For example, if the dataset is in the eu multi-region, then you could specify the URL for the europe-west1 region endpoint, https://europe-west1-aiplatform.googleapis.com/v1/projects/myproject/locations/europe-west1/publishers/google/models/<target_model>. If you specify the model name instead of the endpoint URL, BigQuery ML defaults to using the europe-west4 endpoint for datasets in the eu multi-region, and to using the us-central1 endpoint for datasets in the us multi-region.

Global endpoint

For supported Gemini models, you can specify the global endpoint.

The global endpoint covers the entire world and provides higher availability and reliability than a single region. Using the global endpoint for your requests can improve overall availability while reducing resource exhausted (429) errors, which occur when you exceed your quota for a regional endpoint. If you want to use Gemini 2.0+ in a region where it isn't available, you can avoid migrating your data to a different region by using the global endpoint instead. You can only use a model deployed to the global endpoint with the AI.GENERATE_TEXT function.

Don't use the global endpoint if you have requirements for the data processing location, because when you use the global endpoint, you can't control or know the region where your processing requests are handled.

Processing locations for Google models and partner models

For information about processing locations used by Google models hosted in Agent Platform, see ML processing for Google Cloud models. This information covers models deployed to regions or multi-regions. Models that use the global endpoint don't guarantee any particular processing location.

For information about processing locations used by partner models hosted in Agent Platform, see ML processing for Google Cloud partner models.

Locations for non-remote models

This section contains information about supported locations for models other than remote models, and about where model processing occurs.

Regional locations

The following table contains information about supported locations for all model types other than remote models:

	Region description	Region name	Imported models	Built-in model training	DNN/Autoencoder/ Boosted Tree/ Wide-and-Deep models training	AutoML model training	Hyperparameter tuning	Vertex AI Model Registry integration
Americas
	Columbus, Ohio	`us-east5`	●	●
	Dallas	`us-south1`	●	●
	Iowa	`us-central1`	●	●	●	●	●	●
	Las Vegas	`us-west4`	●	●		●		●
	Los Angeles	`us-west2`	●	●	●			●
	Mexico	`northamerica-south1`	●	●
	Montréal	`northamerica-northeast1`	●	●	●	●	●	●
	Northern Virginia	`us-east4`	●	●	●	●	●	●
	Oregon	`us-west1`	●	●	●		●	●
	Salt Lake City	`us-west3`	●	●	●
	São Paulo	`southamerica-east1`	●	●	●	●
	Santiago	`southamerica-west1`	●	●
	South Carolina	`us-east1`	●	●	●		●	●
	Toronto	`northamerica-northeast2`	●	●		●
Europe
	Belgium	`europe-west1`	●	●	●	●	●	●
	Berlin	`europe-west10`	●	●
	Finland	`europe-north1`	●	●	●
	Frankfurt	`europe-west3`	●	●	●	●	●	●
	London	`europe-west2`	●	●	●	●	●	●
	Madrid	`europe-southwest1`	●	●
	Milan	`europe-west8`	●	●
	Netherlands	`europe-west4`	●	●	●	●	●	●
	Paris	`europe-west9`	●	●
	Stockholm	`europe-north2`	●	●
	Turin	`europe-west12`		●
	Warsaw	`europe-central2`	●	●
	Zürich	`europe-west6`	●	●	●	●	●	●
Asia Pacific
	Bangkok	`asia-southeast3`	●	●
	Delhi	`asia-south2`	●	●
	Hong Kong	`asia-east2`	●	●	●	●	●	●
	Jakarta	`asia-southeast2`	●	●				●
	Melbourne	`australia-southeast2`	●	●
	Mumbai	`asia-south1`	●	●	●	●		●
	Osaka	`asia-northeast2`	●	●	●
	Seoul	`asia-northeast3`	●	●	●	●	●	●
	Singapore	`asia-southeast1`	●	●	●	●	●	●
	Sydney	`australia-southeast1`	●	●	●	●	●	●
	Taiwan	`asia-east1`	●	●	●	●	●	●
	Tokyo	`asia-northeast1`	●	●	●	●	●	●
Middle East
	Dammam	`me-central2`		●
	Doha	`me-central1`		●
	Tel Aviv	`me-west1`	●	●