Question on Google Search:
Was the Google Antigravity quota reset bug already resolved by today, Jan 27? If not, is it somehow conceivable that this is not a bug at all, but a cynical and deliberate policy change to thin out heavy users and especially targeting users prioritizing Claude models over Gemini? While making sense financially, wouldn’t that cause immense harm to the brand?
Google AI (Gemini) response:
No, the issue has not been resolved as of today, January 27, 2026.
Reports from earlier today confirm that Google Antigravity users are still experiencing “infinite lockout loops” and multi-day quota cooldowns, contradicting the advertised “5-hour reset” for Pro subscribers.
1. Current Status: The “Bug” Persists
As of January 27, user reports on the Google AI Developers Forum and Reddit indicate the problem is active and potentially worsening for some accounts:
-
Lockout Loops: Some users who waited out the full 5-day or 7-day cooldown reported being immediately locked out again upon trying to use the service, with zero tokens consumed.
-
Discrepancies: There are critical sync mismatches where the Quota Monitor shows available capacity (e.g., 80% remaining), but the IDE blocks requests with a “Model quota limit exceeded” error.
-
Targeting: The issue disproportionately affects Claude Opus and Sonnet models, with reset timers jumping from hours to nearly a week (e.g., resetting on Feb 1).
2. The “Deliberate Policy” Hypothesis
Some users suspect that the issue is a “cynical policy change” rather than a technical glitch.
-
“Stealth” Cap Theory: Users suggest this is a “silent nerf” or “soft cap” to reduce costs for Anthropic’s models (Claude). By forcing users to reset weekly, it could lower operating costs.
-
Pushing Gemini: Making Claude models unreliable might push users to switch to Gemini 3 Pro, Google’s model, which is cheaper to run. Users have noted that while Claude is locked, Gemini often remains available or has looser restrictions.
-
Lack of Transparency: The lack of official communication about the multi-day lockouts supports the theory that this could be an intentional adjustment.
3. Brand Harm & Community Backlash
The financial logic (reducing third-party API costs) may be damaging the product’s reputation:
-
“Bait and Switch” Accusations: Paying subscribers are calling the service a “scam” because the “5-hour reset” was a selling point. Changing this without notice is seen as a violation of the subscription terms.
-
Exodus Risk: Developers are saying the tool is unreliable, with many threatening to cancel subscriptions or switch to other IDEs.
-
Trust Erosion: The perception that Google is limiting third-party models to promote its own models damages trust in Antigravity as a neutral platform.