Related ToolsChatgptClaude

Google Adds Notebooks to Gemini for Persistent Project Context

Google DeepMind
Image: Google

NotebookLM has had project-based organization since 2023. Now Google is bringing a similar concept to Gemini, its main AI chatbot.

The new notebooks feature lets you group files, past chat conversations, and custom instructions into a single workspace. Open a notebook and start chatting, and Gemini draws on everything inside it as context - no re-explaining your project from scratch each session.

The practical uses are straightforward: an ongoing client account, a recurring research topic, a product launch with dozens of reference docs. Store everything once, and every future conversation in that notebook already knows the background. Custom instructions per notebook also mean you can set a specific tone or persona for different projects without touching your global settings.

How It Stacks Up Against the Competition

This brings Gemini closer to where Claude Projects and ChatGPT Projects have been for months. Both already let users create persistent workspaces with attached files and instructions. Google is closing the gap rather than opening one.

The more interesting question is what this means for NotebookLM. That product - built around deep document analysis, audio overviews, and source-grounded Q&A - now has a functional sibling inside Gemini. Google describes them differently: NotebookLM is for serious document research, Gemini notebooks are a lighter organization layer for everyday chatbot use. But for someone who uploads a few PDFs and wants ongoing AI assistance with them, the distinction is going to blur quickly.

Google hasn't said whether notebooks will be free or locked behind Gemini Advanced (the $20/month paid tier). That detail matters. If notebooks are free, it's a meaningful upgrade for casual users. If they're paywalled, it's another reason to pay - or another reason to just use Claude's free Projects tier instead.

Announced April 9, 2026. No rollout timeline given for when all users will see the feature.