Google and Partner models and generative AI features on Vertex AI areexposed as specificregional endpoints and a globalendpoint. Global endpoints cover the entire world and provide higheravailability and reliability than single regions.

Global endpoint

Selecting a global endpoint for your requests can improve overall availabilitywhile reducing resource exhausted (429) errors. Don't use the global endpoint ifyou have ML processing requirements, because you can't control or know whichregion your ML processing requests are sent to when a request is made.

Supported models

Usage of the global endpoint is supported for the following Google models inspecified regions. For details about which models support the global endpoint,see theGlobal tab in theGoogle model endpoint locations table.

• Gemini 3.1 Pro (Preview)
• Gemini 3 Flash (Preview)
• Gemini 3 Pro (Preview)
• Gemini 3 Pro Image (Preview)
• Gemini 2.5 Pro
• Gemini 2.5 Flash (Preview)
• Gemini 2.5 Flash-Lite (Preview)
• Gemini 2.5 Flash Image
• Gemini 2.5 Flash
• Gemini 2.5 Flash-Lite
• Gemini 2.0 Flash
• Gemini 2.0 Flash-Lite

For information about global endpoint availability for partner models, see theGlobal tab in theGoogle Cloud partner model endpoint locations table.

Use the global endpoint

To use the global endpoint, exclude the location from the endpoint name andconfigure the location of the resource toglobal. For example, the followingis global endpoint URL:

https://aiplatform.googleapis.com/v1/projects/test-project/locations/global/publishers/google/models/gemini-2.0-flash-001:generateContent

For theGoogle Gen AI SDK, create a client that uses theglobal location:

# google-genai >= 0.8.0 is requiredclient=genai.Client(vertexai=True,project='PROJECT_ID',location='global')

For theVertex AI SDK for Python,initialize the SDK using theglobal location:

# google-cloud-aiplatform >= 1.79.0 is requiredimportvertexaifromvertexai.generative_modelsimportGenerativeModelvertexai.init(project='PROJECT_ID',location='global')

Limitations

The following capabilities are not available when using the global endpoint:

• Tuning
• Batch prediction for Anthropic and OpenMaaS models
• Retrieval-augmented generation (RAG) corpus (RAG requests are supported)

Usage of the global endpoint with Provisioned Throughput isavailable only for the following models:

Google model endpoint locations

Google models in Vertex AI are available for the following endpoints:

Google Cloud partner model endpoint locations

Google serves requests from the region that you specified. For some models,Google also offers a global endpoint to improve overall availability and reduceerror rates. The global endpoint can have a separate set of quotas from theregional endpoint and doesn't support data residency requirements. For moreinformation, see the "Regional and global endpoint" section inVertex AI partner models forMaaS.

Partner model endpoints for Generative AI on Vertex AI are available inthe following regions:

Google serves requests from the region that you specified. For some models,Google also offers a global endpoint to improve overall availability and reduceerror rates. The global endpoint can have a separate set of quotas from theregional endpoint and doesn't support data residency requirements. For moreinformation, see the "Regional and global endpoint" section inVertex AI open models forMaaS.

Open model endpoints for Generative AI on Vertex AI are available inthe following regions:

What's next

• Learn more about Generative AI on Vertex AIdataresidency.
• Learn aboutGoogle Cloud regions.
• Learn more aboutsecurity controls byfeature.
• Learn about the models that provide Generative AI on Vertex AI support. SeeGenerative AI foundational modelreference.
• Learn aboutVertex AI locations.