Persistent 429 RESOURCE_EXHAUSTED on Antigravity despite active Google AI Pro subscription and full quota

Hello,

I’ve been experiencing a persistent HTTP 429 RESOURCE_EXHAUSTED error in Antigravity since September 21, 2026 (about a week now), across multiple new conversations, different models (Gemini and Claude), and even after fully signing out/in and restarting the app.

Account: ab***************@gmail.com
Plan: Google AI Pro (5TB), active since August 11, 2026, confirmed in Settings > Account and in one.google.com
Model Quota page (Settings > Models): all models (Gemini 3.6 Flash, Gemini 3.1 Pro, Claude Sonnet 4.6, Claude Opus 4.6, GPT-OSS 120B) show full quota bars, refreshing every ~5 hours.

Steps tried:

  • New conversation with just “test” or “hello” — still 429
  • Signed out and back in with the correct account
  • Restarted Antigravity and the computer
  • Tried different models (Gemini and Claude) — same error on all

Error log:
Error: HTTP 429 Too Many Requests
{
“error”: {
“code”: 429,
“message”: “Resource has been exhausted (e.g. check quota).”,
“status”: “RESOURCE_EXHAUSTED”
}
}

This matches other reports on this forum (e.g. threads about “429 despite quota showing available”). Could you please check my account/backend routing? I rely on Antigravity for my small business website and this has completely blocked my work for a week.

Thank you.

Hi @Abdoulahi_Faye, Welcome to the Forum!

Thanks for reporting this and sharing your logs. It looks like your local token cache got out of sync with our backend. When this happens, the system accidentally enforces the free-tier rate limits even if your Settings page shows full Pro quota.

Since you’ve already tried basic restarts and sign-outs, you’ll need to do a hard reset of your local session state to force a fresh handshake. Here’s how to fix it:

  1. Visit https://one.google.com/settings to confirm that your Google AI Pro subscription is active and not on billing hold or pending cancellation.
  2. Open Antigravity IDE, open the Command Palette (Ctrl+Shift+P on Windows/Linux or Cmd+Shift+P on macOS), and run Antigravity: Reset Onboarding. This completely purges local workbench token cache and reloads the window.
  3. Complete the browser OAuth sign-in flow with your account to re-issue fresh entitlement claims.
  4. Ensure you are on the latest Antigravity release by selecting Help > Check for Updates... or downloading from Google Antigravity - Releases .

For more information on model rate limits, please refer to our official FAQ at FAQ | Google Antigravity Docs.