Tier 1 Postpay project hit "prepayment credits are depleted" 429 for ~18 hours, self-resolved without any action

Gemini API
1

Hi, posting this to add another data point to the reports about the Tier 1 Postpay → “prepayment credits depleted” 429 bug

Setup:

  • Gemini API used via google-genai Python SDK (client.aio.models.generate_content)

  • Models: gemini-2.5-pro / flash variants

  • Billing: Postpay (pay-as-you-go), no prior issues

Timeline (KST):

  • 2026-06-09 21:41 – First 429 RESOURCE_EXHAUSTED:

    {'error': {'code': 429, 'message': 'Your prepayment credits are depleted. Please go to AI Studio at https://ai.studio/projects to manage your project and billing...', 'status': 'RESOURCE_EXHAUSTED'}}

  • 2026-06-09 21:42 ~ 2026-06-10 15:52 – Every single hourly request (ran every hour during market hours) failed with the identical error message, including after exponential-backoff retries (up to 5 retries, 2s/4s/8s/16s/32s).

  • 2026-06-10 15:52 ~ 2026-06-11 09:35 – No requests made (outside our scheduled window).

  • 2026-06-11 09:35 – First request after the gap succeeded, and everything has been normal since.

Key points:

  • We made zero changes on our end (no key rotation, no billing changes, nothing — I was traveling and had no access to touch the system).

  • The error persisted continuously for ~18 hours across a midnight boundary, so it wasn’t a daily quota reset.

  • It then resolved itself with no action from us, consistent with the “Google-side billing sync bug” pattern described in other threads.

Would be great to get confirmation from the Gemini team on whether this was the same sync issue, and whether there’s a way to detect/avoid it (e.g., a status endpoint we could poll) so we don’t burn ~18 hours of degraded service next time.

An 18-hour outage on a paid, postpay-enabled project with zero visibility or alerting from Google’s side is a serious reliability concern for anything running unattended (e.g. automated trading/analysis pipelines).

2

Hello @jasonj52 ,
We have fielded this to the internal team for further action and investigation

3

Hi @jasonj52 ,

Sincere apologies for this. Engineering has confirmed you were affected by an incident / bug which caused your account to be switched to prepay. We detected this and switched it back to postpay.

Again, apologies, this should not happen again in the future.