Google Transforms NotebookLM Into a True ‘Agentic’ Research Companion with Gemini 3.5 & Antigravity

Google’s standout research companion just received its most significant evolutionary leap since launch. Moving far beyond simple document summarization, Google has officially overhauled NotebookLM with powerful agentic capabilities, advanced multi-step reasoning, and an isolated coding environment designed to tackle the most complex data analysis workflows.

The upgrade shifts NotebookLM from a passive reading assistant into an active research partner that writes code, discovers sources, and builds project repositories from scratch.

The features are rolling out globally on the web right now.

The Big Launch: What’s New and Exciting

1. Finally: An Agentic Research Companion

Previously, NotebookLM was most effective if you already possessed a curated stack of documents. Now, you can jumpstart a project using nothing but loose ideas and incomplete questions.

Finally: An agentic research companion

You can now start a notebook by just entering your loose ideas and questions into the chat. NotebookLM will then guide you through building out your repository, even finding and adding (with your permission!) relevant sources from the web.… pic.twitter.com/1h4s25cVOM

— NotebookLM (@NotebookLM) June 8, 2026

The new agentic framework guides you through building a comprehensive source repository directly inside the chat window. With your explicit permission, NotebookLM will autonomously crawl the live web, unearthing primary materials, cross-language perspectives, and highly targeted reference works to anchor your project.

2. Upgraded Chat Powered by Gemini 3.5 and Antigravity

Want a closer look at today’s launch? Here is a breakdown of what’s new and exciting 🧵:

First up: An upgraded, more thoughtful chat experience.

Powered by Gemini 3.5 and @Antigravity, you will now have better visibility into the AI’s thinking process. Plus, each notebook has a… pic.twitter.com/qhZMJwSRDV

— NotebookLM (@NotebookLM) June 8, 2026

The underlying intelligence of NotebookLM has been officially upgraded to a powerhouse combination of Gemini 3.5 and Google’s Antigravity agentic platform. This integration introduces two major user upgrades:

• Visible Thinking Process: You will now have step-by-step visibility into the AI’s logic directly in the chat interface, making answers completely transparent.

• A Secure Cloud Computer: Every single notebook now acts as an isolated virtual machine. NotebookLM can autonomously write and execute Python code in the background to handle heavy data calculations, mathematical modeling, and dataset normalization.

Unlocking this workspace is a suite of over 100+ curated software skills powered by Antigravity, giving the model predefined technical blueprints to execute system-level operations flawlessly.

Ready-to-Download, Customizable Output Formats

Next: You can now ask NotebookLM to assemble context from your sources into ready-to-download, customizable formats. New outputs include:

📊 Data visualizations & charts
📁 PDFs, docx, & markdown files
🖼️ Images with Nano Banana
📑 Excel, PPTX, & Structured data (csv, json) pic.twitter.com/NJWVxhyqyn

— NotebookLM (@NotebookLM) June 8, 2026

Instead of forcing users to manually copy and paste text out of a chat box, NotebookLM can now assemble context directly from your source material into polished, customizable, and production-ready files:

• Data Visualizations & Charts: Export native data analytics as clean .png or vector .svg graphics.

• Documents: Instantly compile your work into professional PDFs, .docx files, Markdown text, or raw .txt summaries.

• Images with Nano Banana: Generate custom contextual imagery, slides, or infographics directly from text sources (supports .png, .jpg, and .gif).

• Structured Data: Turn document logs into clean developer formats like .csv or .json.

• Microsoft Suites: Native creation of complex Excel spreadsheets (.xlsx) with computed math and fully formatted PowerPoint slide decks (.pptx).

Global Rollout Details: Who Gets It First?

Google is deploying these features sequentially across its premium tiers, but emphasizes a long-term strategy for broader access:

According to Google’s internal benchmarking, this Gemini 3.5 architecture achieves an average win rate of over 65% against prior models, spiking to a 78.2% win rate on web-based source discovery. If you have access to a premium Google tier, the update is live on your web dashboard starting today.

Interested in reading more about Google related AI news, rumors and deals. Read our full Google AI news coverage by clicking here.

Please follow us on our Facebook page and X account for all latest and breaking Google, Android and Nokia related news.