Paid Tier 1 project, but Gemini API api_requests quota returns 0 (429)

Gemini API
1

Hello Google Gemini API Team,

I am experiencing a Gemini API quota issue with a Paid Tier 1 project.

Project information:

  • Project ID: gen-lang-client-00715427..
  • Project Number: 1544002806..
  • Billing/Tier: Tier 1
  • Model: gemini-3.1-flash-image (Nano Banana 2)

AI Studio currently shows the following limits for Nano Banana 2:

  • RPM: 100
  • TPM: 200K
  • RPD: 1K

However, actual Gemini API requests consistently fail with HTTP 429 RESOURCE_EXHAUSTED.

The important part of the error is:

quota_metric:

generativelanguage.googleapis.com/api_requests

quota_limit_value:
0

quota_location:
asia-east1

The same quota error also occurs with a simple GET request to:

/v1beta/models

Therefore, this does not appear to be related to image generation, the prompt, or the specific image-generation endpoint.

The project number returned in the API error is also:
1544002806..

This is the same project number associated with the Tier 1 project shown in AI Studio.

Expected behavior:
The project is Tier 1 and AI Studio shows a non-zero quota for Nano Banana 2.

Actual behavior:
The API appears to apply an api_requests quota of 0 in asia-east1, resulting in HTTP 429.

Could you please check whether the api_requests quota for project 1544002806.. is incorrectly provisioned or synchronized?

In particular, could you verify why:

  • AI Studio shows Tier 1 with valid model quotas
  • but the Gemini API returns api_requests quota_limit_value=0
  • and the request is routed to asia-east1?

I have not exceeded the displayed model limits.

Thank you.

2

Hello Google Gemini API Team,

I am following up because the issue is still reproducible on the same project.

Current project status:

  • Billing / Tier: Paid Tier 1

  • Model: gemini-3.1-flash-image (Nano Banana 2)

  • AI Studio displayed limit: 100 RPM / 1,000 RPD

  • Current usage shown in the console: approximately 0% at the time of inspection

  • The project number returned by the API error matches the Paid Tier 1 project shown in Google Cloud Console.

A fresh production request still fails immediately with:

HTTP 429
RESOURCE_EXHAUSTED
RATE_LIMIT_EXCEEDED

The provider error message states:

“Quota exceeded for quota metric ‘API requests’ and limit ‘Request limit per minute for a region’ of service ‘generativelanguage.googleapis.com’.”

This is occurring despite the project being confirmed as Paid Tier 1 and the model-specific quota showing 100 RPM / 1,000 RPD.

We also verified that this is not caused by our application billing or usage accounting. The request reaches the Gemini API and Gemini itself returns the 429 response.

Could you please investigate the backend quota provisioning for this project, particularly the regional generativelanguage.googleapis.com/api_requests quota?

The earlier behavior indicated that this quota may have been provisioned with an effective limit of 0 for a region even though the project is Paid Tier 1.

Could you please verify:

  1. Whether the project’s regional api_requests quota is correctly provisioned for Paid Tier 1.

  2. Whether an effective quota of 0 is still being applied to any region.

  3. Whether the project requires a backend quota/tier synchronization or reprovisioning.

  4. Whether gemini-3.1-flash-image is receiving the correct Paid Tier entitlement.

I can provide the full Project Number privately to a Google team member if required.

For security, I will not post the API key publicly.

Thank you.