The Circuitry
THE CIRCUITRYYour one-stop source for all tech news
HOMENEWSFEEDEVENTS
—STORIES—VERIFIED
BOOKMARKS
RSSSOURCESABOUTCORRECTIONS
RSS
© 2026 The Circuitry
About UsContactCorrections
  • Home
  • Feed
  • Events
  • Saved
Scroll for more
Verification
VERIFIEDConfidence: HIGH
Source identified
Claims cross-referenced
No discrepancies found
Fact-check summary

9to5Google, The Verge, and other outlets confirm NotebookLM's Gemini 3.5 and Antigravity upgrade rolling out June 8, 2026.

Sourcing
1source

via Ars Technica

Ars Technica · track record
16Stories
100%Verified
1330d
All sources →
Markets
GOOGL···

Live quote · not investment advice

Home/Tech
VERIFIEDBy Xavier Rivera· ·2.5 min read

Gemini 3.5 and Antigravity arrive in NotebookLM

Google is updating NotebookLM with the Gemini 3.5 model, Antigravity-powered code execution, multi-format document generation, and automated web source import. The changes are available first to AI Ultra and qualifying enterprise users.

Source:Ars Technica
Post
Gemini 3.5 and Antigravity arrive in NotebookLM
TL;DRAI · 60 sec read

Google rolls out Gemini 3.5 to NotebookLM and integrates Antigravity for running code with over 100 skills. The update supports new output types like PDFs, Excel files, and images. NotebookLM can discover and add web sources from user chats. The changes deliver better accuracy and research capabilities but reach premium accounts first.

Google is rolling out one of the largest updates yet to NotebookLM, moving the AI research tool to the Gemini 3.5 model with expanded capabilities for code execution, document generation, and web sourcing.

NotebookLM upgrades to Gemini 3.5 model. The update replaces the prior Gemini 3.1 branch and delivers improvements across Google's top five core evaluation dimensions: Accuracy and Quality, Multilingual Support, Large Document Analysis, Document Creation, and Advanced Research. In side-by-side tests, the upgraded NotebookLM achieved a 65 percent win rate against the older version. Google conducted the evaluations internally but provided limited details on the specific test methodology.
In side-by-side tests, the upgraded NotebookLM achieved a 65 percent win rate against the older version.

Gemini 3.5 Flash first debuted at Google I/O this year with claims of faster and more efficient processing that can reduce token costs for companies while maintaining or improving output quality. NotebookLM, which launched in 2023, has relied on Google's latest AI models to let users analyze specific documents and webpages.

Antigravity integration adds code execution. NotebookLM now includes its own cloud computer powered by Antigravity, enabling it to write and run code to support research goals. The feature ships with more than 100 software skills that allow users to build workflows inside notebooks without switching between separate applications. These skills expand what NotebookLM can accomplish with user queries.

Expanded output formats beyond text. The tool can now generate documents in multiple formats that appear in the Studio Panel alongside existing outputs like infographics, quizzes, and audio overviews. Supported formats include data visualizations and charts as PNG or SVG, PDFs, DOCX, Markdown, text files, images via Nano Banana as PNG, JPG or GIF, structured data as CSV or JSON, Microsoft Excel XLSX, and Microsoft PowerPoint PPTX. Users can prompt NotebookLM to edit these generated files after creation, and Google plans to add more file types over time.
NotebookLM now includes its own cloud computer powered by Antigravity, enabling it to write and run code to support research goals.

Streamlined web source discovery. NotebookLM can now find and import additional web sources directly from the chat interface. Users ask Gemini to locate relevant context, after which it produces a research report offering the option to import all or selected sources. All future interactions within that notebook incorporate both the manually provided sources and the newly imported ones. This reduces reliance on users supplying every source upfront.

Availability limited to premium accounts initially. The Gemini 3.5 upgrade and associated features begin rolling out today but are restricted to AI Ultra subscribers, Workspace business customers with AI Ultra Access, and those with AI Expanded Access. Most NotebookLM users on standard Google accounts will receive the updates in the near future. The company has not disclosed a specific timeline for broader availability.
Why this mattersAI · ~100 words

Tap a lens to see what this story means for you.

Reader-supported
DonateBuy me a coffee →Follow@thecircuitry_ →

Reader-supported · Daily Brief

Daily brief at 7 AM ET. Top tech stories, every morning. Sourced and fact-checked.

HELP US IMPROVE

Reader-supported

The Circuitry is a passion project I've always wanted to build, and I love the work behind it.

Running it costs real money. APIs, hosting, time. To keep improving the site and growing this into something useful for everyone, those costs have to be covered.

Any contribution is appreciated. If not, no pressure. Thanks for reading.

Buy me a coffee
GoogleNotebookLMGeminiAI
More fromArs Technica
  • Microsoft packages laced with credential stealer for second time

    Tech · 1h
  • S&P 500 Rejects Fast-Track for SpaceX and AI Firms

    Tech · 3d
  • Waymo to Repurpose Robotaxi Batteries for Grid Storage

    Energy · 4d
More inTech
  • Apple Overhauls Image Playground With Photorealistic AI

    Tech · 1h
  • visionOS 27 developer beta 1 rolls out after WWDC26

    Tech · 1h
  • Microsoft packages laced with credential stealer for second time

    Tech · 1h
SupportThe Work

The Circuitry is reader-supported. If you find the daily brief useful, you can buy me a coffee to keep it going.

Buy a coffee →
SubscribeCircuitry Brief

Daily brief at 7 AM ET. Top tech stories, every morning.

MORE IN TECH

Apple Overhauls Image Playground With Photorealistic AI

Apple has announced a major overhaul of Image Playground that adds photorealistic image generation via a new model on Private Cloud Compute, plus expanded editing tools and iOS integration. The update includes automatic SynthID watermarks on all AI images and opens the features to third-party developers through a new API.

visionOS 27 developer beta 1 rolls out after WWDC26

Apple released the first developer beta of visionOS 27 following its announcement at WWDC26. The update brings the Siri AI app, Visual Intelligence features, and system reliability improvements.

Microsoft packages laced with credential stealer for second time

Dozens of Microsoft open source packages were compromised with a credential-stealing payload that activates in AI coding agents, marking the second such supply-chain attack in weeks. The incident highlights weaknesses in the trust model of modern development workflows that rely on cryptographically signed provenance.