Google is giving NotebookLM one of its biggest upgrades yet, adding Gemini 3.5, code execution capabilities, web access, and a range of new tools designed to make complex research projects easier to complete.
Quick Summary – TLDR:
- NotebookLM now runs on Gemini 3.5, bringing improved reasoning and more reliable responses.
- Users can now run code inside a secure cloud computer built into every notebook.
- The platform can generate PDFs, spreadsheets, presentations, charts, and images directly from research materials.
- NotebookLM can now use Google Search to find and suggest sources, helping users start projects with simple ideas instead of preloaded documents.
What Happened?
Google has announced a major upgrade to NotebookLM, its AI-powered research and note taking tool. The update introduces Gemini 3.5 and Google’s Antigravity technology, enabling more advanced reasoning, agentic capabilities, and deeper analysis.
The company says the new features are designed to help users tackle complex research tasks, organize information more efficiently, and transform source material into useful outputs such as reports, presentations, and data visualizations.
Want a closer look at today’s launch? Here is a breakdown of what’s new and exciting 🧵:
— NotebookLM (@NotebookLM) June 8, 2026
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
Gemini 3.5 Brings Smarter Research Capabilities
NotebookLM has evolved significantly since its launch as an experimental project from Google Labs. Originally designed to help users better understand documents and organize information, the platform has become a popular research companion for students, professionals, and organizations.
With the move to Gemini 3.5, NotebookLM is gaining stronger reasoning abilities and more accurate responses. Google says users will also get better visibility into how the system arrives at answers, making the research process more transparent.
The company reported notable improvements during internal evaluations. According to Google, the upgraded NotebookLM achieved an average win rate of more than 65 percent across its key evaluation categories. The system also delivered a 69.9 percent win rate in large document analysis and a 78.2 percent win rate in web research and source discovery compared to the previous version.
Secure Cloud Computer Enables Code Execution
One of the most significant additions is a secure cloud computer attached to every notebook.
This allows NotebookLM to write and execute code directly within the platform. Google says the feature can help users perform advanced analysis, process data, and uncover deeper insights from their sources without needing separate software tools.
The upgrade is supported by more than 100 curated software skills, giving NotebookLM access to a broader range of analytical capabilities.
For researchers working with large datasets, technical documentation, or complex reports, this could dramatically reduce the time required to gather and interpret information.
New Output Formats Expand Productivity
Google is also expanding the types of content NotebookLM can generate.
Users can now create and download a wide variety of files directly from the platform, including:
- PDF reports with charts and tables
- Microsoft Word documents
- Excel spreadsheets
- PowerPoint presentations
- CSV and JSON files
- Markdown and text documents
- Charts and data visualizations
- Images generated through Nano Banana
Another useful addition is the ability to edit outputs after they have been generated, providing more flexibility for users preparing professional reports or presentations.
Web Access Makes Research Easier
Previously, NotebookLM worked best when users supplied their own documents, PDFs, websites, or videos as source material.
Now, users can begin with nothing more than a question or rough idea.
NotebookLM can use Google Search to discover relevant web sources, suggest materials for research, identify related works, and even surface primary sources in different languages. Users remain in control of which sources are added to their notebooks, while attribution continues to be clearly displayed.
This enhancement helps transform NotebookLM from a document analysis tool into a more complete research assistant.
Designed for Researchers, Professionals and Businesses
Google highlighted several practical applications for the new features.
Researchers can combine information from multiple datasets, run analysis through code, and generate charts and reports. Technical professionals can simplify dense documentation into guides and presentations. Small business owners can compare sales data against advertising performance to make more informed decisions about growth and expansion.
The update also reinforces NotebookLM’s unique position in the AI market. Unlike many chatbots that generate responses from general internet knowledge, NotebookLM focuses on source grounded answers, helping users trace information back to the materials being used.
SQ Magazine Takeaway
I think this is one of the most important NotebookLM updates since the product launched. Google is moving beyond simple document summaries and turning NotebookLM into a true AI research workspace. The combination of Gemini 3.5, code execution, web access, and file generation tools makes it far more useful for serious research and professional work. If Google continues expanding these capabilities, NotebookLM could become one of the strongest productivity tools in the AI space.
