Persistent HTTP 429 “Quota exceeded” on Gemini Batch API (Tier 1) — Please restore Tier 1 entitlement

I’m on Tier 1 for the Gemini API, and my usage is within the published Tier 1 limits for models/gemini-2.5-flash. However, for the past 3 days, every attempt to create a Batch job fails immediately with HTTP 429 RESOURCE_EXHAUSTED (“You exceeded your current quota…”).

Why this is unexpected

Tier 1 limit shown for my account is 3,000,000 tokens (Tier 1 policy).

My Batch submission is far below this threshold (each batch is ~10,000 input tokens; the error occurs on the very first batch creation).

This suggests my account/project is being treated as having zero quota, incorrect tier assignment, or another quota/entitlement issue unrelated to actual token volume.

Error (exact)

RuntimeError: HTTP 429 {
“error”: {
“code”: 429,
“message”: "You exceeded your current quota, please check your plan and billing details. For more information on this error, head to: https://ai.google.dev/gemini-api/docs/rate-limits. To monitor your current usage, head to: https://ai.dev/rate-limit. ",
“status”: “RESOURCE_EXHAUSTED”
}
}

Request
Please investigate and fix whatever is causing the API to return 429 despite Tier 1 compliance, so I can use my Tier 1 quota normally. Specifically, please confirm:

My account/project is correctly recognized as Tier 1 for models/gemini-2.5-flash

Batch creation is not blocked by an unexpected quota dimension (e.g., batch-job creation limits, project-level quota, billing state mismatch)

Any quota/billing entitlements have been correctly applied and are active

Details to help you reproduce / diagnose

Model: models/gemini-2.5-flash

API: Batch API (job creation step)

Behavior: Fails on the first batch creation request every time

Duration: ~3 days continuously

I can provide: project ID, region, request IDs, timestamps, and my rate-limit dashboard screenshot upon request

Thank you. I’d appreciate a quick confirmation of the root cause and the steps needed to restore Tier 1 access.

Hi @Marcin_Lee,

There was an issue with Batch API earlier and it was fixed now. To troubleshoot your issue, please provide a screenshot of your usage details from https://ai.dev/usage?tab=rate-limit and separately DM your project number.

Thank you!