News listGoogle admits fault after community backlash: Antigravity Gemini rate limits increased by 3x, weekly quota reset
動區 BlockTempo2026-05-22 02:11:59

Google admits fault after community backlash: Antigravity Gemini rate limits increased by 3x, weekly quota reset

ORIGINAL社群炮轟後 Google 認錯:Antigravity Gemini 速率限制提高 3 倍、重置本週額度
AI Impact AnalysisGrok analyzing...
📄Full Article· Automatically extracted by trafilaturaGemini 翻譯1604 words
Varun Mohan, a core member of Google Antigravity, publicly admitted to previous decision-making errors and announced a permanent 3x increase in Gemini rate limits for all paid tiers of Antigravity, while immediately resetting the weekly quota for all users. (Previous coverage: Google launches Antigravity 2.0: Gemini 3.5 Flash engine, dynamic sub-agents, and scheduled features are now live) (Background: Google shifts promotion criteria to "AI performance," employees fragment product ecosystem) The founder of Windsurf, who was acquired by Google last year along with about 40 core engineers for a $2.4 billion licensing fee, is currently a core member of Antigravity. Paid users drove Antigravity's ratings to an all-time low, and then Varun Mohan stepped up to admit the mistake. Yesterday afternoon, he stated on X: "We made mistakes in some of our decisions." An update: we’re 3xing the rate limits for Gemini models across all paid tiers in Antigravity and resetting everyone’s Gemini quota for the week. We understand some people hit their rate limits quickly and wanted to respond fast. Lots more to come and enjoy building! — Varun Mohan (@_mohansolo) May 21, 2026 Google recently significantly adjusted the usage limits for Gemini, shifting from "daily query counts" to "compute-based" calculations. The system now consumes quotas based on prompt complexity, the model used, features, and conversation length. Furthermore, the reset cycle has been changed to every 5 hours, though there is still a total weekly cap. This means that even if you haven't exhausted your Antigravity weekly quota, performing too many computations in a short period will still result in temporary rate limiting (resetting every 5 hours). The quota issue with Antigravity is the result of four months of consecutive tightening, not something that suddenly erupted after I/O 2026. The timeline is clear: - December 2025: Free tier Gemini quota significantly reduced. - February 2026: Image generation quota tightened. - March: A point-based system introduced for the subscription-based billing model. - After the I/O conference in May, Antigravity 2.0 went live, and even $250/month Ultra paid users were subjected to rate limiting. From the user's perspective, paying the same monthly fee for increasingly fewer capabilities naturally leads to dissatisfaction. The period after I/O 2026 was the breaking point, with developers posting monthly bills on Reddit and X, calculating the gap in token consumption for the same tasks between the old and new versions of Antigravity. Mohan's concessions consist of two parts: First, a permanent 3x increase in Gemini model Rate Limits for all paid tiers. Varun Mohan emphasized in the announcement that this is not a temporary adjustment, but a permanent policy change. Second, an immediate reset of all users' weekly quotas, without waiting for the natural reset next week. This allows users to immediately feel that "something has returned" rather than just receiving empty promises. However, a 3x quota increase does not solve the deeper issue: the transparency of the pricing logic. Developers want to know why token consumption for the same tasks surges with version updates. Will the quota standards be adjusted again? How far in advance will the team notify the community before the next reduction? Varun Mohan's announcement did not directly answer these questions; the 3x quota increase is more of an emotional fire-extinguishing measure than a systemic reform. In 2026, the AI software development tool market is exceptionally competitive. With Claude Code, Cursor, GitHub Copilot, and Continue all present, the market offers more choices and lower switching costs. Any large-scale user dissatisfaction could accelerate churn that is difficult to recover from.
Data Status✓ Full text extractedRead Original (動區 BlockTempo)
🔍Historical Similar Events· Keyword + Asset Matching6 items
💡 Currently matching via keywords + symbols (MVP) · Will be upgraded to embedding semantic search later
Raw Information
ID:ef3334b3c7
Source:動區 BlockTempo
Published:2026-05-22 02:11:59
Category:zh_news · Export Category zh
Symbols:Unspecified
Community Votes:+0 /0 · ⭐ 0 Important · 💬 0 Comments