Google has announced that Gemini Notebook will replace its fixed daily feature caps with a compute-based usage model beginning September 2, 2026, a structural shift that changes how every consumer account on the platform tracks and spends its AI allowance.
Fixed daily caps give way to a compute budget
Google is moving Gemini Notebook usage limits to a compute-based system, the same approach it introduced for the Gemini app in May.
The changes had not applied to Gemini Notebook, which continued offering prompt-based limits ranging from three audio and video overviews and ten reports or quizzes per day on the free tier to as many as 5,000 daily chat queries and 200 audio and video overviews on the Ultra plan.
Under the new system, a user’s overall usage limit will factor in prompt complexity, chat length, number of sources, and the features you use. A short factual query costs the compute budget far less than generating a multi-source Video Overview or a full Slide Deck.
This update, affecting both web and mobile platforms, rolls out to all consumer accounts on September 2.
Five-hour refresh window and weekly cap replace daily reset
Because limits refresh every five hours until a weekly cap is reached, users can keep working throughout the day. If a user hits their usage limits, they can come back later as the quota refreshes every five hours until they reach their weekly limit, or upgrade Gemini Notebook to get higher AI limits.
To help users navigate the change, Google has integrated a live usage indicator directly below the chat window, providing immediate feedback on how much of the current five-hour capacity has been consumed.
Studio generations will carry a separate bar showing expected AI usage cost. The more the bar is filled in, the higher the expected cost.
The notebook will suggest alternative outputs if a user’s first choice surpasses the limit, and users can defer generating outputs such as Video Overviews or Slide Decks for later. Google will automatically create those deferred outputs once the limit resets, with this feature coming first to the web.
Tiered multipliers scale with subscription plan
On the AI Plus plan users get two times higher usage than standard limits, while the Pro plan bumps it to four times. With AI Ultra, the usage limit goes up to five times or twenty times higher than AI Pro, depending on the subscription.
The shift mirrors the tiered compute model Google rolled out for the main Gemini app earlier this year. It signals a broader platform strategy of metering AI workloads by actual compute cost rather than by a flat count of requests.
That approach is increasingly common across the industry as the server cost of running large AI inference models at scale makes undifferentiated flat limits difficult to sustain. Earlier this summer, an AI coding tool silently had its context cap cut by 27 percent overnight. Separately, Google expanded Gemini access to college students with a free year of AI Pro, making the new limit structure directly relevant to that growing user base.
What changes on September 2
The new compute-based usage limits will start rolling out to consumer accounts on web and mobile on September 2. The rollout applies to all subscription tiers.
Gemini Notebook usage limits are designed to ensure an optimal experience for everyone, which means Google may at times cap the number of prompts and conversations users can have, or how much they can use some features, within a specific timeframe.
The company acknowledged in its official support page that Gemini Notebook limits may change as the system matures.
For heavy users accustomed to batching large multi-source research sessions into a single daily window, the five-hour rolling refresh, combined with a weekly ceiling, represents a meaningful operational change. Spreading intensive work across the day will now make better use of the available compute budget than queuing it all at once.
