Google adjusts Gemini’s new usage limits in response to complaints
Google is changing Gemini’s compute-based limits after users complained about hitting caps too fast, and it will add clearer usage breakdowns.
Intelligence analysis by GPT-5.4 Mini
Google is tweaking Gemini’s new compute-based quota system after complaints that limits were being hit too quickly. The company also says failed requests will not count, Flash-Lite is now free, and more usage details are coming.
Google changed the rules for Gemini because some people ran out of uses too fast. Instead of counting every request the same way, it now looks at how hard the job is, like how a truck uses more fuel for a heavy load than for a small box.
If a request fails, it does not use up the allowance. Google also says it will show clearer notes about where the uses went, so people can understand what used the most.
It also fixed a video bug and made a lighter version free. That means some tasks should feel less like a game with a tiny timer and more like a fair meter that matches the work being done.
Analysis
What changed
Google switched the Gemini app to a compute-based usage system at I/O 2026, and that change drew complaints about users hitting limits too quickly. In response, the company says it is adjusting the limits and the way quota is measured.
The new system is meant to account for how expensive a prompt is to run. Google says it considers prompt complexity, the tools used, and chat length, so a simple text prompt should cost less than a coding task or a video-heavy request. Josh Woodward said Google is now capping how much quota a single prompt can consume in Gemini 3.1 Pro, which should help large prompts and big files avoid burning through a user’s allowance so quickly.
Google also clarified a few points that make the system less punishing. If a request fails, it will not count against quota, and successful completions are the only ones that use quota. Heavy tasks like Deep Research will still use more compute, but Google says it will provide more detailed usage breakdowns and notifications so people can see where their limits are going. The current usage dashboard only offers a high-level view.
Smaller changes and fixes
Google says 3.1 Flash-Lite prompts are now free and will not count against quota. It also says the app will remember a user’s chosen model across sessions unless the user changes it or Gemini automatically falls back to a lighter model after a cap is reached.
There was also a bug affecting some users where just one or two Omni videos could drain quota quickly. Google says it fixed that issue, and Gemini AI Ultra users now get double the number of Omni generations.
The company says it plans to offer pay-as-you-go top-up AI credits in the future, which suggests the current quota model may eventually get a more flexible paid option.
Key points
- Google changed Gemini to a compute-based quota system and is now adjusting it after user complaints.
- Single prompts in Gemini 3.1 Pro will have a cap on how much quota they can consume.
- Failed requests will not count against quota, and Google plans more detailed usage breakdowns.
- Gemini 3.1 Flash-Lite prompts are now free and do not count toward quota.
- Google fixed an Omni video quota bug and doubled Omni generations for AI Ultra users.
