AI News HubLIVE
站内改写3 min read

Do your best research with NotebookLM

NotebookLM, Google's AI research assistant, receives major upgrades including advanced reasoning, new output formats (PDFs, spreadsheets, etc.), and easier research initiation. Built on Gemini 3.5 and Antigravity, it offers improved accuracy and analysis capabilities.

SourceHacker News AIAuthor: xnx

Do better research with NotebookLM

We’re introducing across the board upgrades to NotebookLM that deliver new agentic capabilities and more advanced reasoning to tackle complex research projects.

Your browser does not support the audio element.

Listen to article

This content is generated by Google AI. Generative AI is experimental

[[duration]] minutes

Three years ago we launched NotebookLM as an experimental AI product from Google Labs to help you understand anything. Millions of people and organizations turn to NotebookLM as a collaborative knowledge and research partner because it helps them organize their thinking, identify deeper connections across their documents and spark new ideas. Today we’re introducing across the board upgrades to NotebookLM that deliver new agentic capabilities in chat and more advanced reasoning to tackle the most complex research projects.

An upgraded, more thoughtful chat experience

First, we’re upgrading NotebookLM to run on Gemini 3.5 and Antigravity providing even more accurate and reliable information along with better visibility into the thinking process.

Each notebook is now equipped with a secure cloud computer, enabling NotebookLM to write and run code useful for helping you perform deeper research and more complex analysis. The system includes more than 100 curated software skills, unlocking a wide range of new capabilities to help you more deeply understand the sources in your notebook.

In our side-by-side evaluations against our prior system, the upgraded NotebookLM achieved an average win rate of over 65% (a 15% point margin above parity) across our top five core evaluation dimensions. The system demonstrated substantial improvements in large document analysis, securing a 69.9% win rate and achieved exceptional performance in advanced web research and source discovery reaching a 78.2% win rate against our prior baseline.

1

More output formats, with improved customization

You can now ask NotebookLM to create outputs in even more formats and provide detailed instructions to guide the outputs. NotebookLM will assemble the context from your sources to produce useful and high quality formats like PDF reports with charts and tables, detailed budget spreadsheets, and bespoke student worksheets — all ready to download directly from the studio panel. You can even make edits after your outputs are generated. New output formats include:

Data visualizations and charts (png, svg)

Documents (PDFs, docx, markdown, text files)

Images with Nano Banana (png, jpg, gif)

Structured data (csv, json)

Microsoft Excel (xlsx)

Microsoft PowerPoint (pptx)

We plan to add additional formats in the future to help make NotebookLM more helpful in more ways.

More effortless research

Finally, we’re making it easier to get your research started. Previously, NotebookLM was most helpful when you came with your own sources and a good handle on your project. You can now start with loose ideas and questions. NotebookLM can guide you through building your source repository directly in your chat, making it even easier to get started on any project. Perhaps you want to find primary sources in other languages to better understand new perspectives, or you’re seeking related works by an author you recently discovered. It can even use Google Search to find relevant, high quality sources from the web and add them to your notebook.

You’re still in control of the sources that you add to your notebook so that your work stays grounded in the information you trust and all sources continue to be clearly attributed.

More helpful in more situations

Here are some workflows these improvements enable:

Researchers: A data analyst can combine data from various countries with conflicting formatting. To make this information useful, the analyst can ask NotebookLM to conduct web research to find additional context, write code to perform accurate data analysis, and create charts and a PDF report to showcase the results.

Technical Professionals: A program manager can decipher complex specifications for customer integration, instantly transforming the technical documentation into a polished, simplified guide, slide deck and a step-by-step roadmap for the team.

Small Business Owners: A gym owner can run a media campaign and analyze raw sales data against ad spend. By calculating the campaign's financial impact using NotebookLM, the business owner is better equipped to decide whether to expand to other cities.

These updates are rolling out globally on the web starting today to all users with Google AI Ultra and all Workspace business customers with AI Ultra Access and AI Expanded Access, and we plan to expand to others over time. We look forward to your feedback.

You can also subscribe with a

.

POSTED IN:

1

Evaluation set contains queries spanning core NotebookLM use cases across source-grounded Q&A, multilingual interactions, long-form document understanding, content generation, and multi-source research. Category definitions: Accuracy & Quality evaluates the correctness, relevance, and groundedness of responses against user-uploaded sources; Multilingual Support measures the ability to understand queries and generate faithful responses across non-English languages; Large Document Analysis assesses reasoning and comprehension over long-context source material such as books, reports, and transcripts; Artifact Creation evaluates the ability to recognize when a user query warrants generating a structured artifact such as a summary, study guide, FAQ, or briefing document; Web Research measures the ability to discover, retrieve, and synthesize sources from the web to augment notebook context.

NotebookLM

Dive deeper into I/O 2026 with NotebookLM.

Arts & Culture

Explore Benjamin Franklin’s scientific discoveries on NotebookLM

Mar 30, 2026

NotebookLM

Generate your own Cinematic Video Overviews in NotebookLM.

Mar 04, 2026

Google Labs

Organize your insights with Data Tables in NotebookLM.

Google Labs

NotebookLM adds Deep Research and support for more source types

Nov 13, 2025

Google Labs

Create flashcards and quizzes in the NotebookLM app.