Gemini Notebook Usage Limits Change September 2
Google is replacing daily feature caps with a compute-based system that refreshes every five hours and varies by task complexity.
Gemini Notebook usage limits are changing on September 2, when Google will replace set daily feature caps with a compute-based system that accounts for the complexity of prompts, chat length, number of sources, and features used.
Gemini Notebook usage limits will work differently
The new system replaces fixed numbers of daily generations that varied by Google AI tier. Instead, Gemini Notebook will track a shared usage allowance based on how much compute a task is expected to require. Limits refresh every five hours, continuing until the weekly limit is reached.
Google says the notebook will show usage information inside the chat, including when the current limit will reset. Studio generations will also display an indicator for expected AI usage cost, with a fuller bar representing a higher expected cost.
The change also introduces a fallback for heavier requests. Gemini Notebook can suggest alternative outputs when the preferred option would exceed the available allowance. For resource-intensive outputs such as Video Overviews and Slide Decks, a new "Generate later" option can defer the generation until the limit refreshes. Google says this feature is initially available on the web.
Paid plans get higher limits
Google's new limits scale with its consumer AI subscriptions:
- Without a plan: Standard limits.
- Google AI Plus: 2x higher than standard limits.
- Google AI Pro: 4x higher than standard limits.
- Google AI Ultra: 5x or 20x higher than AI Pro, depending on the subscription.
The change brings Gemini Notebook closer to the compute-based model Google introduced for the Gemini app earlier this year. Saganote previously covered Google's rename of NotebookLM to Gemini Notebook, while the current Gemini pricing guide covers the broader differences between Google's AI subscription tiers.
Google says the new compute-based limits will begin rolling out to consumer accounts on the web and mobile starting September 2. The company describes the change as a way to give users more control over how their compute allowance is spent, while the practical limit for any task will depend on its complexity.