Vertex AI API returning 429 error despite 0% usage and valid quota

Google Cloud Build with AI Generative AI & Foundational Models
1

My project (PII Removed by Staff) is unable to make any successful calls to the Vertex AI API for the imagegeneration@006 model in the us-central1 region.

Every single API request fails with a 429 RESOURCE_EXHAUSTED error, even after days of inactivity.

We have confirmed the following:

  • The quota for imagegeneration in us-central1 has been approved and is set to 65 requests/minute.

  • The project has an active and linked billing account.

  • The Vertex AI API and Cloud Resource Manager API are enabled.

  • Local authentication credentials have been correctly set with gcloud auth application-default set-quota-project.

The project is behaving as if the effective quota is zero, despite the console showing a limit of 65. The issue is not with the client code but seems to be an internal provisioning problem with the project itself. Please investigate the backend logs to see why our approved quota is not being honored.

4

Hey Bruce,
The imagegeneration@006 model is being deprecated and will be removed on September 24, 2025 . Image generation API  |  Generative AI on Vertex AI  |  Google Cloud

I recommend migrating to a newer Imagen model since that is just a couple of weeks away.