Gemini Notebook

Gemini Notebook is Google's standalone source-grounded research notebook, renamed from NotebookLM on July 16, 2026. Google explicitly describes it as the same product, now more integrated with the Gemini app and Google Search. More than 30 million people and 600,000 organizations were using it at the time of the rename. ¹

The core workflow remains notebook-first: users assemble sources, ask questions against them and receive answers with inline source citations. Standard access supports up to 100 notebooks, 50 sources per notebook and 50 chats per day; higher access tiers raise those limits substantially. The product also generates reports, flashcards, quizzes, mind maps, Audio Overviews and Video Overviews, and includes Deep Research for web-based source discovery. ²

The 2026 product line extends beyond summarization. Google added a secure cloud-computer layer so notebooks can execute code for deeper data analysis, and the product syncs with notebooks inside Gemini. Independent coverage from Ars Technica, TechCrunch, The Verge and Engadget corroborates the same product transition and advanced analysis/output expansion. ¹³⁴⁵⁶

Standard Gemini Notebook access remains free. Google uses access tiers rather than a simple Notebook-only subscription: eligible Google AI and Workspace plans unlock higher notebook, source, chat, Deep Research and overview limits, with the highest tiers reaching 500 notebooks, up to 600 sources per notebook and much larger daily generation limits. Because these entitlements are bundled into broader Google plans and vary by account/region, the access tier is more reliable than quoting one universal Notebook-only paid price. ²

The free tier is unusually capable for serious research: 100 notebooks, 50 sources per notebook, 50 daily chats, three daily Audio Overviews and three daily Video Overviews are documented for standard access. The trade-off is that advanced code execution and the highest generation quotas may arrive first or exclusively on higher Google AI/Workspace tiers. ²¹

The evidence base is materially stronger than for many AI research tools. Current exact-product App Store and Google Play listings both show 4.8/5 ratings, with roughly 5.6K UK App Store ratings and about 291K Google Play reviews. These ratings describe the mobile apps, so they are useful evidence about the mobile experience rather than direct proof of the full web research surface. ⁷⁸

Editorial coverage is also broad. TechCrunch and Ars Technica documented the Gemini 3.5/Antigravity research upgrade; The Verge and Engadget covered the July 2026 rename and secure code-execution direction; Tom's Guide reported hands-on use of Deep Research; WIRED found Audio Overview customization useful but variable for narrowly constrained prompts. These sources support capability and usability context, not transferable model benchmark scores. ^4356910

The product's trust advantage is architectural rather than absolute: answers are designed to stay grounded in selected sources and expose inline citations, which makes verification easier than with a generic chatbot. That does not make generated synthesis infallible; users should still inspect the cited passages, especially for high-stakes research or generated multimedia. Mobile reviews also indicate that organization and feature parity can lag the web experience. ⁷

Gemini Notebook is particularly strong for students, researchers, analysts and other knowledge workers who already have documents to understand but also want the system to discover additional web sources. It works well when provenance matters because the notebook keeps sources visible and answers cited. Users who mainly want open-ended chat, autonomous actions across arbitrary business systems or a stable developer API will find it less suitable.