Global and multi-regional endpoints

Cloud Translation - Advanced offers a global endpoint as well as EU and US multi-regional endpoints:

  • (global)

If you use a multi-regional endpoint, your data at-rest and machine learning processing stays within the continental boundaries of the EU or US. These multi-regional endpoints are important if your data's location must be controlled to comply with local regulatory requirements.

If you don't specify an endpoint, Cloud Translation - Advanced uses the global endpoint by default.

Global versus multi-regional endpoints

When using a multi-regional endpoint, there are some difference when compared to using the global endpoint:

  • Text translations that use custom AutoML models are not supported. You can use only the pre-trained NMT model.
  • Translating formatted documents (also known as document translations) is not supported
  • Features that are not GA (still in Preview) are not supported.
  • Calls through the global endpoint cannot access resources that were created by using a multi-regional endpoint. Similarly, calls through a multi-regional endpoint cannot access resources that were created by using the global endpoint.

Restricting resource locations

Organization policy administrators can restrict the regions available for Cloud Translation - Advanced resources by creating a resource locations constraint. If set, Cloud Translation - Advanced users would be able to create resources only in a particular location.

Specify an endpoint

The following example shows a text translation that uses a multi-regional endpoint. If you use the client libraries, set the API endpoint as part of the client options. For some examples, see Setting the location using client libraries in the Cloud Natural Language API guide.


Before using any of the request data, make the following replacements:

  • PROJECT_NUMBER_OR_ID: the numeric or alphanumeric ID of your Google Cloud project
  • ENDPOINT: Regional endpoint, which determines where your data resides. For example,
  • LOCATION: Region where you want to run this operation. You must pick a region inside the continental boundary of the regional endpoint. For example, if you use the endpoint, specify a region in Europe such as europe-west1.

HTTP method and URL:

POST https://ENDPOINT/v3/projects/PROJECT_NUMBER_OR_ID/locations/LOCATION:translateText

Request JSON body:

  "model": "projects/PROJECT_NUMBER_OR_ID/locations/LOCATION/models/general/base",
  "sourceLanguageCode": "en",
  "targetLanguageCode": "de",
  "contents": ["Come here!"]

To send your request, choose one of these options:


Save the request body in a file named request.json, and execute the following command:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "x-goog-user-project: PROJECT_NUMBER_OR_ID" \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \


Save the request body in a file named request.json, and execute the following command:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred"; "x-goog-user-project" = "PROJECT_NUMBER_OR_ID" }

Invoke-WebRequest `
-Method POST `
-Headers $headers `
-ContentType: "application/json; charset=utf-8" `
-InFile request.json `
-Uri "https://ENDPOINT/v3/projects/PROJECT_NUMBER_OR_ID/locations/LOCATION:translateText" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

  "translations": [
      "translatedText": "Komm her!",
      "model": "projects/PROJECT_NUMBER_OR_ID/locations/LOCATION/models/general/base"