Google Upgrades NotebookLM With Antigravity to Run Code and Analysis

Google Upgrades NotebookLM With Antigravity to Run Code and Analysis

N
News Editor 01
2026-07-23 03:35:14
Google has upgraded NotebookLM with Gemini 3.5 Flash and built-in Antigravity support, giving the research tool the ability to write code and run analysis inside a cloud computer environment.
GoogleNotebookLMGeminiAntigravityAI research tools

Google has rolled out its biggest NotebookLM update so far, replacing the underlying model with Gemini 3.5 Flash and adding built-in support for Antigravity. The change pushes NotebookLM beyond document Q&A and note organization, giving it the ability to write code, execute programs, and handle analysis tasks inside the product.

Gemini 3.5 Flash becomes the new core model

The engine swap is a major part of the release. Google said Gemini 3.5 Flash was introduced at Google I/O 2026, runs 4 times faster than other frontier models, outperforms Gemini 3.1 Pro on nearly all benchmarks, and can lower token costs for enterprises that move projects onto Flash.

Google also shared internal comparison results between the new and previous NotebookLM versions. In side-by-side testing across five areas—accuracy and quality, multilingual support, large file analysis, file creation, and advanced research—the Gemini 3.5 version posted an average 65% win rate. The company did not provide much detail on the methodology, so outside verification is still limited.

NotebookLM now gains agent-style execution

The more consequential shift is the integration of Antigravity. Google introduced Antigravity at I/O 2026 as an agent-first AI development platform, and version 2.0 now spans desktop apps, CLI, and SDK. It is designed for AI agents that can plan tasks, write code, open browsers for testing, fix errors, and deploy software.

Google has also folded the standalone Gemini CLI into Antigravity CLI, pointing to a broader consolidation of its agent tooling. Inside NotebookLM, that translates into a “secure cloud computer” attached to each notebook, allowing the system to generate and execute code directly within the context of a user’s research instead of only suggesting code snippets.

Broader file support and easier source collection

Google said NotebookLM now supports more file formats and simplifies the process of bringing in web sources. A user can start with loose ideas or early-stage questions, then have NotebookLM use Google Search to gather higher-quality online sources and build a source library automatically.

That expands the product’s role. Rather than serving only as a tool that answers questions about uploaded material, NotebookLM is being positioned to cover a larger workflow that includes source discovery, organization, analysis, and content generation.

Available first to AI Ultra and select Workspace customers

The upgraded NotebookLM is being released first to Google AI Ultra subscribers and Workspace business customers with AI Ultra access. Users on other plans will need to wait for later availability.

Google is clearly pushing agent-based AI into everyday research use, not just developer environments. How well that works in practice will depend on the stability of Antigravity during complex tasks, the kinds of errors it produces, and whether non-technical users can actually rely on those execution features once the update is in wider use.

This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan.
300

Disclaimer:

The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.

Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.