Tier 1 paid project blocked by free-tier quota (limit 0) on gemini-3.1-flash-image — matches long-running pattern in this category

Gemini API
, ,
1

Project: PlacdAI MVP
Project Number: 4147671294..
Model: gemini-3.1-flash-image (“Nano Banana 2”), called via direct generateContent, Gemini Developer API (not Vertex AI)

Error:

Quota exceeded for metric: generativelanguage.googleapis.com/generate_content_free_tier_requests, limit: 0, model: gemini-3.1-flash-image
Quota exceeded for metric: generativelanguage.googleapis.com/generate_content_free_tier_input_token_count, limit: 0, model: gemini-3.1-flash-image

What I’ve already ruled out, to save a round-trip on the usual checks:

  • Billing account is Paid, not Free Trial — confirmed via Billing Overview (real spend history, “Paid account” label, no free-trial flag)
  • Both API keys show “Tier 1 · Prepay” directly in AI Studio, under this project
  • Manual cash prepay on file (not just pay-as-you-go) — added Aug 8
  • Rate Limits dashboard shows this exact model correctly assigned non-zero Tier 1 limits: 3/100 RPM, 2.57K/200K TPM, 11/1K RPD — nowhere near breached before this started
  • A brand-new key created after the error started reproduces the identical error
  • Text models on the same project (gemini-3.5-flash, etc.) work fine under paid limits — only this image model is affected
  • Checked IAM & Admin → Quotas & System Limits in Cloud Console — Generative Language API doesn’t expose quotas there, so no override to find/remove on that path

Started: Aug 9, 2026. Still ongoing.

Also have a reopened Google Cloud Billing case (#74225195) if useful for cross-reference — billing support confirmed my account is correctly Paid/Tier 1 but couldn’t action anything since it’s a quota-routing issue, not a billing one.

Given how long this exact pattern has persisted across this category (over a year, multiple models, several “critical bug” threads), I know a single new report isn’t going to move the needle much on its own — but wanted the specific model/error combo on record in case it helps narrow down what’s changing between models when this fires. On a launch timeline this week, so genuinely appreciate any visibility from the team.

2

Hi,
We were unable to reproduce this issue, would it be possible to share if you are experiencing this for all image models or only this specific model itself.

3

I’m hitting this exact issue. Billing is active (Tier 1 · Prepay) on my project, confirmed via the “Setup complete” linking flow, but gemini-3.1-flash-image still returns:
429 RESOURCE_EXHAUSTED — generate_content_free_tier_requests, limit: 0

To rule out a project-specific glitch, I created a second, brand-new project and linked the same billing account to it — identical error, same model, same limit:0. This has now persisted over 24 hours across two separately-provisioned projects, so it doesn’t look like a one-off sync delay.

Project IDs: gen-lang-client-0275920174 and gen-lang-client-0151804612. Any idea what’s actually gating this?