Chat in NotebookLM: A powerful, goal-focused AI research partner

We're rolling out changes to NotebookLM to make it fundamentally smarter and more powerful. First, a set of back-end improvements provide a major boost to performance and quality, and second, we are expanding the ability to set goals for your notebook, allowing each notebook to adapt to your needs. Here's what's new.

Get deeper insights and clearer answers

We have made fundamental upgrades to how chat in NotebookLM works. These back-end improvements, powered by the latest Gemini models, work together to improve performance, quality and contextual understanding. Since we started testing these improvements, we've seen a 50% improvement in user satisfaction with responses that use larger amounts of sources. Here's everything we've improved:

• More seamless and natural conversations. We have significantly expanded NotebookLM's processing capabilities, conversation context and history. Starting today, we're enabling the full 1 million token context window of Gemini in NotebookLM chat across all plans, significantly improving our performance when analyzing large document collections. Plus, we've increased our capacity for multiturn conversation more than sixfold, so you can get more coherent and relevant results over extended interactions.

• Deeper insights. We have enhanced how NotebookLM finds information in your sources. To help you uncover new connections, it now automatically explores your sources from multiple angles, going beyond your initial prompt to synthesize findings into a single, more nuanced response. This is especially important for very large notebooks, where careful context engineering is critical in delivering a high quality and trustworthy answer, grounded on the most relevant information in your sources.
• Saved and secure conversation history. To support long-term projects, your conversations will now be automatically saved. You can now close a session and resume it later without losing your conversation history. You can delete chat history at any time, and in shared notebooks, your chat is visible only to you. This will start rolling out to users over the next week.

Personalize your chat with goals

Now available to all users, you can customize chat to adopt a specific goal, voice or role - ranging from a PhD student analyzing sources to a creative storyteller exploring ideas. To get started, click the configuration icon in the chat.

Then, write how you want your chat to behave and the goal you want to achieve. Here are a few to try:

• Treat me like a PhD candidate: You are my research advisor. Rigorously challenge every assumption. Ask probing questions, identify logical fallacies and force me to defend my work from the ground up.
• Act as a lead marketing strategist: Your response must be an immediate action plan. Be analytical and direct, and focus exclusively on the concrete strategies and critical-path steps needed to achieve the goal fast.
• Analyze the provided material from three distinct perspectives: As a strict academic focusing on evidence and logical consistency, as a creative strategist looking for non-obvious connections and innovative applications, and as a skeptical reviewer actively searching for gaps, flaws and potential problems in the conclusions.
• Act as a Game Master for a text-based simulation. Present a high-stakes scenario with a specific goal and a step limit (e.g., 10). I make all the choices. Narrate the outcomes vividly, using realistic, scenario-relevant details.

By integrating our latest model improvements, we're delivering more precise, insightful, and context-aware responses than ever before. With the ability for everyone to set goals for conversions, NotebookLM is now better equipped to adapt to your specific project. We hope these changes unlock new levels of productivity and creativity for your work.

• Google Labs
• AI
• Gemini Models