Google has tripled Gemini usage limits for Antigravity, twice

The transition to the “Agentic AI era” just hit a massive infrastructure bottleneck. Merely days after Google I/O 2026 introduced strict “compute-used” usage models across its premium subscriptions, Google DeepMind was forced to stage an emergency backpedal.

Following intense backlash from the programming community, Google has permanently tripled the Gemini request limits inside its Antigravity developer platform—twice within a 48-hour window. If you are keeping score, this rapid succession of emergency patches delivers a massive 9x overall increase compared to the highly restrictive baseline caps enacted early last week.

📉 Chronology of a Post-I/O Crisis: Why the Caps Broke

The friction began when Google retired legacy, predictable flat-rate daily prompt counts in favor of an algorithm that measures the processing complexity of multi-step agents. Under this new backend system, every automated file search, terminal execution, and sequential codebase scan consumes “compute tokens.”

 [ Early Last Week ]  ──► Restrictive Compute Caps Enacted ──► Developers locked out within an hour.
 [ Mid-Week Patch ]    ──► First Emergency 3x Limit Bump     ──► Codebases still draining quotas rapidly.
 [ End-of-Week Patch ] ──► Second Consecutive 3x Limit Bump  ──► Cumulative 9x cap expansion deployed.

Because long-horizon agentic workflows require continuous, sequential model evaluations across massive multi-file codebases, developers ran into a brick wall. Power users reported exhausting their entire weekly premium allowances within a single hour of active coding, triggering aggressive five-hour cooling-off lockouts on tools they were actively paying to use.

🛠️ DeepMind Responds: Moving from a 3x to a 9x Cushion

As developer forums and Reddit subthreads exploded with accusations of a corporate bait-and-switch, Varun Mohan—a Director within Google DeepMind working directly on the Antigravity core team (and former founder of the Windsurf editor)—publicly stepped in on X (formerly Twitter) to handle damage control.

Mohan openly admitted that the engineering group had made critical errors in their initial assumptions regarding resource consumption during long editing loops. The team deployed two distinct, back-to-back mitigation updates directly to premium developer dashboards:

The First Triple (Wednesday): Google pushed an automated notification expanding model rate limits by 3x and re-crediting weekly user quotas to let developers resume paused projects.
The Second Triple (Thursday Night): Realizing that a simple 3x multiplier was still burning out “after a couple of intensive work sessions,” DeepMind tripled the weekly quota limits on top of the previous expansion, compounding the change into a permanent 9x cap increase.

💬 The Dashboard Alert: Premium subscribers are actively seeing a critical banner notification launch within their Antigravity workspaces stating: “If you are on a paid plan, your Gemini quota has been reset for the week and increased… Keep building!”

📊 Structural Summary: The Antigravity Quota Adjustments

Operational Phase	Gemini Limit Framework	Developer Practical Impact
Initial Post-I/O Launch	Strict compute-used usage tracking.	Triggered sudden lockouts within 60 minutes of multi-file editing.
Wednesday Emergency Patch	3x Rate Limit Multiplier Deployed.	Restored access but complex agents still hit the technical wall quickly.
Thursday Emergency Patch	Additional 3x Weekly Quota Multiplier.	Yields a 9x capacity surge; safely stabilizes long-horizon coding workflows.
Global Quota Status	Wiped & Reset Twice Natively	All paid plans re-credited with zero historical consumption penalties.

⚠️ The Fine Print: Restricting the Cushion to Code

While developers are celebrating this emergency correction, the tech industry points out that Google’s concession is highly localized. This massive 9x quota expansion applies strictly to model calls executed within the Antigravity development workspace.

The broader, standard consumer usage limitations controlling the core Gemini application, generative Workspace utilities, and media generation tasks remain tightly bound to the standard compute-caps introduced at the conference.

. Google Search ads generative UI, AI booking tools and search agents

. What’s New in Search

. We’re introducing new ways to design in real time with Stitch.

. I/O ’26 Recap: Everything You Need to Know

. Introducing Gemini Omni: Create Anything from Anything

. TPU Training Day for I/O ‘26

. Google I/O ’26 Keynote

. Meet Gemini Omni

. The Gemini app becomes more agentic, delivering proactive, 24/7 help

. Fuel your next wave of growth on YouTube with Demand Gen

. Everything Google announced at I/O 2026: Gemini, Search, Android XR, & more

OTHER