Google is upgrading its NotebookLM research tool by integrating the Gemini 3.5 model and the Antigravity platform. This rollout expands the AI-powered assistant’s ability to generate a wider variety of outputs and leverages new research to enhance how users interact with and synthesize complex information.
The update represents a significant shift for NotebookLM, moving the tool beyond simple document summarization toward a more robust research environment. By upgrading the underlying models and integrating the Antigravity research framework, Google is aiming to provide users with deeper reasoning capabilities and more sophisticated ways to interact with their data.
As Google continues to iterate on its AI ecosystem, the integration of Gemini 3.5 into specialized tools like NotebookLM highlights a broader trend: moving away from general-purpose chatbots toward highly specialized, research-oriented agents. This upgrade focuses on three specific pillars: the underlying intelligence of the models, the diversity of the outputs produced, and the foundational research made possible by the Antigravity platform.
The shift to Gemini 3.5 intelligence
At the heart of this upgrade is the transition to Gemini 3.5. While the specific technical benchmarks of this model remain part of Google’s ongoing development, the integration into NotebookLM is designed to improve the assistant’s ability to process and reason through large volumes of information. For researchers and students, this means the tool can better grasp context, identify subtle connections between disparate documents, and provide more accurate answers to complex queries.
The move to a more advanced model architecture allows NotebookLM to handle more nuanced instructions. Instead of merely extracting text, the tool is being positioned to synthesize information, allowing users to bridge the gap between raw data and actionable insights. This evolution is critical for users who rely on the tool to manage extensive libraries of PDFs, notes, and research papers.
Integrating the Antigravity platform
A key component of this rollout is the research facilitated by Antigravity. According to recent developer updates, Antigravity serves as a platform that supports new ways of building and testing AI-driven capabilities. In the context of NotebookLM, this platform has provided the research foundation necessary to implement the current suite of upgrades.
The involvement of Antigravity suggests that the improvements in NotebookLM are not just incremental software updates but are the result of new research methodologies. This integration allows Google to push the boundaries of how an AI assistant can assist in the research process, moving from a reactive tool to one that can proactively help structure and interpret information through more advanced computational frameworks.
Expanded outputs and research workflows
Perhaps the most visible change for the end-user will be the expansion of available outputs. The upgrade explicitly mentions a move toward “more outputs,” which indicates that NotebookLM will offer a broader range of ways to transform source material. While the specific new formats are being rolled out, this expansion is expected to include more diverse ways to visualize, structure, and present research findings.
Users can expect the tool to move beyond text-based summaries to provide more varied content types. This could include more structured data representations, different styles of explanatory content, or enhanced ways to interact with the synthesized information. The goal is to allow the user to consume their research in the format that best suits their specific workflow, whether that is preparing a presentation, writing a report, or studying for an exam.
By diversifying the ways in which information is presented, Google is attempting to solve one of the primary friction points in AI-assisted research: the “wall of text” problem. Providing varied outputs helps users move more quickly from the stage of information gathering to the stage of knowledge application.
Official rollout schedules and specific feature availability are expected to be shared through Google’s developer and product update channels in the coming weeks. We will continue to monitor these updates as they become available.
What do you think about the integration of Gemini 3.5 into your research workflow? Share your thoughts in the comments below and share this article with your network.