Google just made NotebookLM a lot more useful with its biggest upgrade yet

Affiliate links on Android Authority may earn us a commission. Learn more.

There aren’t many AI tools I rely on every single day, but NotebookLM is one of them. It has become my go-to tool whenever I need to sift through large amounts of information without losing track of where the facts came from. Now, Google’s latest update aims to make that experience even better.

NotebookLM will now run on Gemini 3.5 and Antigravity, giving the research-focused AI tool its biggest upgrade yet. This should make responses more accurate, improve reliability, and offer better transparency.

Every notebook will now include its own secure cloud-based computing environment. That means the AI won’t just summarize information or answer questions; it can also write and execute code when needed, opening the door to more advanced workflows that previously required jumping between multiple apps.

Google says the upgraded experience includes over 100 specialized software skills and a broader set of tools designed to help users dig deeper into their source material. Instead of simply reading through documents and notes, users can now transform that information into polished outputs in a wide variety of formats.

So, say for instance, you’re researching a topic, you could turn your collected sources into a PDF report. NotebookLM can also generate and export reports in PNG and SVG formats, documents including PDF, DOCX, Markdown, and plain text files, Nano Banana images in PNG, JPG, and GIF formats, structured datasets in CSV and JSON, Excel spreadsheets, and PowerPoint presentations.

What’s particularly interesting is that the system is becoming more multilingual. Users can provide instructions in one language and receive completed work in another, making it more useful for international research projects, cross-border collaboration, or studying sources that aren’t available in your native language.

One of NotebookLM’s biggest limitations has always been its starting point. The tool worked best when you already had a collection of sources and knew exactly what you wanted to research. Google is now trying to remove that barrier. Instead of arriving with a carefully assembled folder of documents, you can now start with nothing more than a question, rough idea, or topic you want to explore. It can help build a source library directly inside the chat experience.