Gemini Notebook usage limits, explained
On September 2, 2026, Gemini Notebook (formerly NotebookLM) switched to compute-based usage limits. Every chat message and generation now draws from a quota that depends on prompt complexity, the model and features you use, and conversation length. Here is what changed, the full limits for every plan, and what to do when you hit the cap.
What changed on September 2, 2026
Gemini Notebook moved from fixed per-feature counters to compute-based usage limits. Instead of every chat or generation costing the same, each interaction now consumes quota based on the complexity of your prompt, the models and features involved, and the length of your chat.
Quota refreshes on a rolling basis — roughly every five hours — until you reach your weekly limit. When you hit the cap, Gemini Notebook tells you exactly when features become available again instead of leaving you guessing.
Limits remain subject to change. Google notes that access can be capped or adjusted for testing, experimentation, or availability — treat the numbers below as a snapshot and check the official pages linked at the end for current values.
How much each plan gets
Under the new system, Google describes paid tiers as multipliers of the standard free allowance rather than fixed counters:
| Plan | AI usage limit |
|---|---|
| Free (no AI plan) | Standard limits |
| Google AI Plus | 2× standard limits |
| Google AI Pro | 4× standard limits |
| Google AI Ultra | 5× or 20× AI Pro, depending on your subscription |
The familiar per-feature quotas still exist alongside the compute budget. The next table lists every published number.
Every limit by plan
Published usage limits as of September 2026. Daily quotas reset after 24 hours; monthly quotas reset after 30 days.
| Feature | Free | AI Plus | AI Pro | AI Ultra (20 TB) | AI Ultra (30 TB) |
|---|---|---|---|---|---|
| Notebooks | 100 | 200 | 500 | 500 | 500 |
| Sources per notebook | 50 | 100 | 300 | 500 | 600 |
| Chat messages | 50/day | 200/day | 500/day | 2,500/day | 5,000/day |
| Audio Overviews | 3/day | 6/day | 20/day | 100/day | 200/day |
| Video Overviews | 3/day | 6/day | 20/day | 100/day | 200/day |
| Cinematic Video Overviews | — | — | 2/day | 10/day | 20/day |
| Reports | 10/day | 20/day | 100/day | 500/day | 1,000/day |
| Flashcards | 10/day | 20/day | 100/day | 500/day | 1,000/day |
| Quizzes | 10/day | 20/day | 100/day | 500/day | 1,000/day |
| Mind Maps | 10/day | 20/day | 100/day | 500/day | 1,000/day |
| Deep Research | 10/month | 3/day | 20/day | 75/day | 200/day |
| Data Tables | Limited | More limits | High limits | Higher limits | Highest limits |
| Infographics | Limited | More limits | High limits | Higher limits | Highest limits |
| Slide Decks & Revisions | Limited | More limits | High limits | Higher limits | Highest limits |
Artifacts Gemini Notebook generates automatically when you first add sources — a report, flashcards, an infographic, a slide deck, or an audio/video overview — are created once and do not count toward these limits. Sharing a notebook does not change the source limit for collaborators; everyone keeps the limit of their own plan.
How the compute budget works
Compute-based limits mean two requests can cost different amounts of quota. Three things drive the cost of each interaction:
- Prompt complexity. A quick question across two sources costs less than synthesizing fifty sources into a structured brief.
- Model and features. Premium capabilities — cinematic video overviews, deep research, higher-tier Gemini models — consume more of the budget per use.
- Conversation length. Longer chats carry more context, so each additional message gets a little more expensive.
Gemini Notebook surfaces the budget in three places:
- Settings → Usage. A panel at the top right shows your current AI usage and weekly limit.
- Chat footer. A message shows remaining usage and the reset time — for example, “You’re almost at your AI usage limit. Limit resets at 3:00 PM.”
- Studio panel. Before you generate an artifact, a bar shows the expected AI usage cost — the fuller the bar, the more of your budget it will use.
What to do when you hit the limit
Running out of quota does not lock the app — you have three options:
- Wait for the refresh. Quota refills in rolling windows of about five hours, up to your weekly cap. The chat footer shows the exact reset time.
- Use Generate later. On the web, Studio artifacts can be queued instead of generated immediately: open the Studio panel, select an artifact, and choose Generate later. The generation may take a couple of hours, and you get a notification when it finishes.
- Upgrade. The Upgrade button in chat or Studio leads to Google AI plans — Plus doubles the standard limits, Pro quadruples them, and Ultra goes 5–20× beyond Pro.
What paid plans add beyond limits
Higher tiers are not only about bigger numbers:
- Model access. Plus and Pro unlock more capable Gemini models; Ultra gets the highest access tier.
- Earlier features. Pro and Ultra get priority access to new capabilities; Plus gets early access ahead of free users.
- Advanced sharing. Paid plans can control how shared notebooks are used. Custom chat and analytics stay available to everyone.
- Watermark removal. Pro and Ultra can toggle off the visible watermark on generated images and videos — except in India, South Korea, and Vietnam, where it is applied automatically.
- Enterprise option. Through Google Cloud, Gemini Notebook for Enterprise offers 5× or more artifact limits plus VPC-SC, IAM controls, and enterprise-grade data protection.
Good to know
- Gemini Notebook is the new name for NotebookLM — same product, same notebooks, same limits.
- A distinct badge next to your profile picture confirms you are on an upgraded plan.
- Workspace and Workspace for Education users may get expanded access through a qualifying edition.
- Generated images and videos carry an invisible SynthID watermark on every plan.
- If a limit looks different from this guide, Google’s pages win — they update first.
Usage limit questions
What are the new Gemini Notebook usage limits?
Since September 2, 2026, Gemini Notebook uses compute-based limits: each chat message or generation consumes quota based on prompt complexity, the model and features used, and conversation length. Quota refreshes roughly every five hours until you reach a weekly cap.
How often do Gemini Notebook limits reset?
The compute quota refreshes on a rolling basis about every five hours, until you hit your weekly limit. Per-feature quotas still apply too — daily quotas reset after 24 hours and monthly quotas after 30 days.
How do I check my Gemini Notebook usage?
Open Settings → Usage at the top right of Gemini Notebook. The chat footer also shows remaining usage and your next reset time, and the Studio panel shows the expected cost before you generate an artifact.
What is Generate later in Gemini Notebook?
When you are out of AI limits on the web, Generate later lets you queue Studio artifacts instead of waiting. In the Studio panel, select an artifact and choose Generate later — it may take a couple of hours, and you are notified when it finishes.
Did NotebookLM limits change too?
NotebookLM is now called Gemini Notebook — it is the same product, so the new compute-based limits apply to everything you used to do in NotebookLM.
Do shared notebooks share usage limits?
No. Sharing a notebook does not change the source limit or quota for collaborators — each person keeps the limits of their own plan.