Tool profile · Research

Gemini Notebook

Formerly NotebookLM, a research workspace grounded in sources you choose

Gemini Notebook is the new name for Google’s NotebookLM. It organizes PDFs, web pages, media, documents, and other material into notebooks, then provides cited answers, synthesis, and generated artifacts grounded in the selected sources.

Source-grounded Q&AResearch assistantDocument analysisAudio OverviewsMind mapsStudy workflows

01 / Overview

About

Google renamed NotebookLM to Gemini Notebook on July 16, 2026. It remains a standalone research product while expanding its connections to the Gemini app and Google Search. The core workflow is unchanged: create a notebook for one project, add or discover sources, select the relevant material, ask questions, compare evidence, and create outputs such as reports, Audio or Video Overviews, mind maps, quizzes, infographics, and slide decks. Some newer capabilities and limits depend on the plan, account type, and rollout status.

02 / Core features

Core features

01

Grounded answers with citations

Chat works from the enabled notebook sources and provides inline citations so readers can return to the original context.

02

Broad source support and discovery

Supports PDFs, web pages, public YouTube videos, audio, Google Docs and Slides, Word, Markdown, images, and more, with source discovery across the web or Drive.

03

Turn sources into useful artifacts

Can create reports, Audio and Video Overviews, mind maps, quizzes, infographics, and slide decks for reading, review, and sharing.

04

Research discovery and expanding data analysis

Fast Research and Deep Research help discover material, while higher plans are progressively gaining code execution and deeper analysis through a secure cloud computer.

03 / Use cases

Use cases

01

Papers, reports, and topic research

Collect multiple sources in one notebook, compare claims, extract evidence, and trace answers back through citations.

02

Courses and exam review

Turn lectures, recordings, and reading material into study guides, Audio Overviews, mind maps, flashcards, and quizzes.

03

Project knowledge and onboarding

Centralize project documentation, meeting records, and context so a team can ask questions from the same source set.

04

Multiple ways to absorb long material

Generate audio, video, infographics, or slides alongside reading and choose the format that fits the situation.

04 / Platforms and languages

Platforms and languages

Platforms and devices

WebiOSAndroid

Languages

Supports 80+ languages

Browser and mobile apps

05 / Pricing and quotas

Pricing and quotas

The free Standard tier covers the core research workflow. Google AI Plus, Pro, and Ultra primarily raise notebook, source, chat, and generation limits. Plan prices vary by region and account type.

Standard

Free

100 notebooks, 50 sources per notebook, 50 chats per day, and 3 Audio Overviews per day.

Plus

See local Google AI plan pricing

200 notebooks, 100 sources per notebook, 200 chats per day, and 6 Audio Overviews per day.

Pro

See local Google AI plan pricing

500 notebooks, 300 sources per notebook, 500 chats per day, and 20 Audio Overviews per day.

06 / Similar products

Similar products

Perplexity

Start with cited search, then continue into file analysis and deeper research

research

ChatGPT

One workspace for questions, writing, research, images, files, and coding

assistant

08 / Common questions

Frequently asked questions

01Are Gemini Notebook and NotebookLM different products?

No. Google announced on July 16, 2026 that NotebookLM was renamed Gemini Notebook and remains a standalone research product.

02How is it different from a general chat assistant?

Gemini Notebook chat is grounded in the selected notebook sources and provides inline citations, making it better suited to research that needs traceable evidence.

03How much material fits in the free tier?

Standard currently allows 100 notebooks and up to 50 sources per notebook. One source can contain up to 500,000 words, and a local upload can be up to 200MB.

04Are uploads used directly to train foundation models?

Google states that notebook files, generated outputs, and chat history are not used directly to train its foundation models unless the user submits feedback; feedback can include related context and may be reviewed.