Continuing the discussion from Handling 429 / 503 errors from the Gemini API:
Subject: Critical API Error: HTTP 429 RESOURCE_EXHAUSTED / Cloud AI Companion Route Block
Hello Google Support Team,
I am encountering a persistent rate-limiting block while using the Google AI platform. My environment is throwing an immediate HTTP 429 error, indicating that resources have been exhausted, despite my usage remaining well within valid project quota boundaries.
The issue appears to be a backend routing or authentication desynchronization originating at the Edge Server Filter (ESF) layer.
Please see the specific trajectory data and trace logs below for your engineering team to investigate:
- Trajectory ID: 9cb04606-f4a7-421c-9acb-67c49018…
- TraceID: 0x9ffa7b132b61..
- X-Cloudaicompanion-Trace-Id: 9ffa7b132b6178..
- Server Gateway: ESF
- Timestamp of Incident: Mon, 07 Sep 2026 06:25:01 GMT
Error Payload:
json
{
"error": {
"code": 429,
"message": "Resource has been exhausted (e.g. check quota).",
"status": "RESOURCE_EXHAUSTED"
}
}
Use code with caution.
Public developer forums suggest this is a known issue regarding project token bucket synchronization. Could you please check the backend routing for this specific TraceID, verify my project’s quota status, and manually flush the token/rate-limit bucket for my account if it has become stuck?
Thank you for your assistance, and I look forward to your update.
Best regards,
Shiva
Hello @shiva_kumar0422 , please note that there is currently no ongoing platform issue or outage regarding 429 RESOURCE_EXHAUSTED errors. To help us investigate your specific case, could you please let us know which model you are calling when this error occurs?
Hi
@Payal_Sharma2,
Thank you for getting back to me.
The issue is still not resolved. I continue to receive the same immediate HTTP 429 RESOURCE_EXHAUSTED error when attempting to access Gemini AI models, even though I am using my Gemini Pro account/subscription and my usage is not exceeding the applicable limits.
Since you mentioned that there is no ongoing platform-wide outage, could you please investigate this as an account/project-specific issue?
The error is consistently returned with:
- HTTP Code: 429
- Status:
RESOURCE_EXHAUSTED
- Message:
Resource has been exhausted (e.g. check quota).
- Trajectory ID:
9cb04606-f4a7-421c-9acb-67c490..…
- Trace ID:
0x9ffa7b132b6..
- X-Cloudaicompanion-Trace-Id:
9ffa7b132b617...
- Server Gateway: ESF
- Timestamp: Mon, 07 Sep 2026 06:25:01 GMT
The important point is that this is not a one-time transient 429. The same error continues to occur, preventing me from accessing the Gemini models.
Regarding your question about the model: I am experiencing this while accessing Gemini AI models through my Pro account. If you need the exact model identifier, please let me know what information you need from my side and I will provide it.
Could you please check whether there is any account-level, project-level, quota, entitlement, or backend rate-limit issue associated with my account/project, and escalate this to the appropriate engineering team if necessary?
At this point, I would appreciate help determining why my requests are being rejected with RESOURCE_EXHAUSTED despite my Pro access and apparently available quota.
Thank you.
Hello, could you please try starting a new chat session in Google AI Studio and check if you are able to send messages there? This will help us determine whether the error is specific to your current chat session or affecting your entire project.
Hi Payal,
Just wanted to update you that the issue has been resolved on my end.
I was previously using the earlier Antigravity version, where it was simply shown as Antigravity. After installing the newer Antigravity IDE v2.5.5, the application is now explicitly identified as Antigravity IDE.
Since upgrading to this newer IDE version, I am no longer getting the HTTP 429 RESOURCE_EXHAUSTED error, and the Gemini/agent functionality is working normally again.
So, in my case, the issue appears to have been resolved after upgrading to the newer Antigravity IDE version.
Thanks for following up on this discussion. Hopefully this information is useful to others who may be experiencing the same issue.