Key Takeaways
How to use NotebookLM with Google Gemini - Tutorial

- NotebookLM is now Gemini Notebook, with each notebook getting its own cloud computer for code execution
- Google Search will integrate third-party apps like Instacart and Canva directly through AI Mode
- The new system outperforms its predecessor 65% of the time, reaching 78.2% for advanced web research
Google is folding NotebookLM into the Gemini brand. The research assistant, now called Gemini Notebook, gains cloud compute capabilities that let each notebook write and run code. Separately, Google Search is opening up to third-party apps, letting users interact with services like Instacart and Canva without leaving the search interface.
What's changing with Gemini Notebook?
According to Google VP Josh Woodward, NotebookLM has grown to 30 million individual users and 600,000 organizations. The rebrand to Gemini Notebook is more than cosmetic. Each notebook now gets its own cloud computer capable of writing and executing code, a significant expansion from the tool's original document Q&A focus.
The code execution feature will roll out first to AI Ultra and Workspace customers, with broader access coming over the following weeks. Google's internal benchmarks show the new system beats its predecessor 65% of the time in head-to-head comparisons. For advanced web research tasks, that advantage climbs to 78.2%.
This fits Google's broader pattern of consolidating its AI products under the Gemini umbrella. Bard became Gemini. Duet AI became Gemini for Workspace. NotebookLM was one of the last standalone AI tools in Google's portfolio without the Gemini name attached.
Why add code execution to a research tool?
NotebookLM launched in 2023 as a way to have conversations with your documents. You upload PDFs, notes, or web pages, and the AI answers questions about them. The Audio Overview feature, which generates podcast-style discussions from uploaded materials, drove much of its popularity.
Code execution changes the tool's scope. Instead of just summarizing and discussing your research, Gemini Notebook can now analyze data, run calculations, and prototype solutions. For product teams working through technical documentation or engineering specs, the notebook becomes a more active collaborator rather than a passive reference.
The 78.2% improvement on advanced web research suggests Google is positioning this as a serious alternative to dedicated research tools. Whether that holds up against Perplexity and other AI search products will depend on how the feature performs outside internal benchmarks.
Disclosure
Some links in this post are affiliate links — Logicity earns a commission if you sign up, at no extra cost to you. We only link products we have used or actively recommend.
Third-party apps are coming to Google Search
The second announcement is unrelated to Gemini Notebook but signals a shift in how Google thinks about Search. Starting this week in the US, users can connect apps like Instacart, Canva, and YouTube Music through AI Mode.
The integration is practical. Search for a recipe, and you can add ingredients directly to an Instacart cart. Look for presentation templates, and Canva surfaces them within Search. Ask for workout music, and YouTube Music builds a playlist. This app connection feature already existed in the standalone Gemini app. Google is now extending it to Search proper.
For developers and product teams, the implication is clear: Google is creating another distribution channel for app functionality. If Search becomes a place where users complete transactions without opening separate apps, being part of that integration layer matters. Google says more partners will follow the initial launch.
What this means for AI product development
Both announcements point in the same direction. Google is collapsing the distance between asking questions and taking action. Gemini Notebook moves from summarizing research to executing code based on it. Search moves from showing links to completing tasks within the interface.
For teams building AI agents or automation workflows, the context layer becomes critical. A notebook that can run code needs accurate source material to produce useful output. Research on enterprise AI agents shows that bad context is the primary failure mode, with 57% of wrong answers traced back to context problems rather than model limitations.
Explains why context quality determines AI agent success, relevant to Gemini Notebook's expanded capabilities
The Search integration also raises questions about automation workflows. Tools like Zapier, Make, and n8n currently handle the connections between apps that Google is now building natively into Search. Whether Google's approach competes with or complements these tools depends on how deep the third-party integrations go.
Logicity's Take
Google is betting that code execution and app integration will differentiate Gemini Notebook from competitors like Perplexity and ChatGPT's document analysis features. The 78.2% win rate on advanced research sounds impressive, but internal benchmarks are marketing numbers until independent testing confirms them. For AI product teams, the more interesting signal is Google treating Search as an action platform, not just an answer engine. If that model works, expect every major AI assistant to follow.
Frequently Asked Questions
When will Gemini Notebook's code execution be available?
Code execution launches first for AI Ultra and Workspace customers, with broader access rolling out over the coming weeks.
Is NotebookLM going away?
NotebookLM is being rebranded as Gemini Notebook. The core functionality remains, with code execution added as a new capability.
Which apps can connect to Google Search?
Initial partners include Instacart, Canva, and YouTube Music. Google says more partners will follow.
How does Gemini Notebook compare to its predecessor?
Google's internal testing shows the new system wins 65% of head-to-head comparisons, rising to 78.2% for advanced web research tasks.
Is the Search app integration available outside the US?
The feature is launching this week in the US only. Google has not announced international availability.
Need Help Implementing This?
Building AI-powered research or automation tools? Our team can help you evaluate how Gemini Notebook and similar platforms fit your product stack. Contact us for a consultation.
Source: The Decoder / Matthias Bastian
Manaal Khan
Tech & Innovation Writer
Produced with AI assistance and reviewed by the Logicity editorial team. Learn more in our Editorial Policy.
Related Articles
More in AI & Machine Learning
Bezos AI Lab Gets $10B: What Project Prometheus Means
Jeff Bezos is closing a $10 billion funding round for Project Prometheus, an AI lab focused on physics-based AI for manufacturing and engineering. With a $38 billion valuation and backing from JPMorgan and BlackRock, this signals a major shift in enterprise AI investment toward industrial applications.

Kimi K2.6 Open-Weight AI: 300 Agents at a Fraction of the Cost
Moonshot AI's Kimi K2.6 matches GPT-5.4 and Claude Opus 4.6 on coding benchmarks while running 300 parallel agents. For businesses locked into expensive API contracts, this open-weight model could slash AI infrastructure costs while delivering enterprise-grade automation.




