Persistent HTTP 429 error for gemini-3.6-flash

Gemini API
1

Hi,

I’m getting a persistent HTTP 429 error when using Google Search grounding with the Gemini API, and I’m trying to understand what I should do next.

My setup:

  • Gemini API (not Vertex AI)
  • Model: gemini-3.6-flash
  • Tier 1 with billing enabled
  • Interactions API

A plain request to gemini-3.6-flash using the same API key and project works correctly:

HTTP 200

But as soon as I enable Google Search grounding:

tools: [{"type":"google_search"}]

I consistently receive:

HTTP 429
too_many_requests
You exceeded your current quota, please check your plan and billing details.

I checked the AI Studio Rate Limit page and no limits appear to be exhausted:

  • Gemini 3.6 Flash RPM: 0 / 1K
  • TPM: 0 / 2M
  • RPD: 0 / 10K
  • Gemini 3 Search grounding: 0 / 1.5K

The Search grounding request fails on the first attempt, so this does not appear to be caused by a burst of requests.

The issue has persisted for more than 48 hours.

I have now submitted the official “Report unexpected 429’s in Gemini API” form and included screenshots of the Rate Limit page.

Could someone advise:

  1. Is there anything else I should check or do while waiting?
  2. Does this look like a Search grounding entitlement/backend issue?
  3. How quickly does Google usually review or respond to reports submitted through the unexpected 429 form?
  4. Should I also open another support request, or is submitting the form enough?

I can provide my Google Cloud Project Number privately to a Google team member if needed.

Thanks!

2

Hello @yura0203 ,

I have DM`ed you for more information .

3

it also happend to me but on antigravity ide, can you help me?

4

Hello @yura0203 ,

Please check that the API key used in your request is associated with your billed Tier 1 project rather than an unbilled Free Tier default project. else create a new api key associated to your paid project.

Grounding with Google Search requires an active paid billing tier. So if your API key was created under a separate default project or before attaching billing account, the key might be getting evaluated against Free Tier limits and hence throwing the error.

5

Thank you so much! That was the error! It works perfectly now.