Hi team,
I am experiencing an entitlement desync issue with Antigravity (both standalone IDE and the VS Code extension).
The Problem:
- I have active Google AI Pro subscriptions across separate accounts/family group members.
- When switching between different accounts, the quotas under “Models & Usage” remain completely stuck with the exact same numbers (e.g. Gemini weekly limit: 93%, Claude/GPT weekly limit: 35%).
- Switching accounts or signing in with clean/incognito browser sessions does not update the limits.
Troubleshooting steps already attempted on macOS:
- Explicitly signed out and cleared browser session tokens.
- Deleted Application Support directory:
rm -rf ~/Library/Application\ Support/Antigravity*
- Cleared local Gemini caches and config:
rm -rf ~/.gemini/antigravity-ide and rm -rf ~/.gemini/config
- Cleared macOS Keychain Access credentials.
- Tested on fresh VS Code extension environment.
Despite all local cache purges, the clients immediately fetch the same stale quota state upon login, which indicates a server-side entitlement desync.
Could you please trigger a manual entitlement resync on the backend for my account?
Thank you!
Hi,
I too faced this same issue today. After some testing, I was able to narrow down the cause: the Antigravity model quota appears to be shared among members of the same Google Family group.
I tested with multiple accounts:
Account A (Google AI Pro) — showed the expected Pro quota.
Account B & C (same Google Family group) — showed the same Pro entitlement and the same exhausted/shared quota.
Account D (outside the Family group) — received a completely separate Starter quota with 100% remaining.
So it appears this is not a local caching or login issue; the quota is being shared at the Google Family level.
Hope this helps others facing the same confusion.
Hello,
I am experiencing a persistent model quota synchronization issue too.
I have an active Google AI Pro subscription. For approximately one week, the Model Quota section in the Antigravity Desktop IDE has continuously shown:
“Refreshes in 4 hours, 59 minutes”
This value appears for all available models, including:
- Gemini 3.6 Flash (High)
- Gemini 3.6 Flash (Medium)
- Gemini 3.6 Flash (Low)
- Gemini 3.1 Pro (High)
- Gemini 3.1 Pro (Low)
- Claude Sonnet 4.6 (Thinking)
- Claude Opus 4.6 (Thinking)
- GPT-OSS 120B (Medium)
The value does not appear to progress or update when I check it again later. It remains at approximately 4 hours 59 minutes.
However, the Antigravity CLI reports that my account has quota available:
Weekly Limit Remaining:
99.22%
Refreshes in 24h 21m
Five Hour Limit Remaining:
100.00%
Quota available
This indicates that my account/subscription has available quota and that the CLI can retrieve the correct quota state, while the Desktop IDE appears to be displaying stale or incorrect quota information.
Environment:
Antigravity version:
1.107.0
Architecture:
x64
Additional troubleshooting already performed:
- Signed out of Antigravity.
- Signed back in with the same Google account.
- Verified that the issue persists after re-authentication.
- Checked the quota through the Antigravity CLI.
- Confirmed that the CLI reports 100% remaining on the Five Hour Limit.
- Confirmed that the Weekly Limit is 99.22% remaining.
The issue has persisted for approximately one week.
There is also a separate problem with the Desktop feedback form: after entering the feedback, the Submit action does not appear/work, so I am unable to report the issue through the IDE itself.
same problem, al cuota used 4,59hrs remains. never used antigravity helpp,
Additional troubleshooting already performed:
- Signed out of Antigravity.
- Signed back in with the same Google account.
- Verified that the issue persists after re-authentication.
- unstall, reinstall
- cloud billing support ai help…
- try to open a case, but not product for ai_studio
