0 readers reading
NotebookLM's New Research Mode: What the Gemini 3.5 Update Actually Does for You

NotebookLM's New Research Mode: What the Gemini 3.5 Update Actually Does for You

NotebookLM's June 2026 update adds Gemini 3.5, a secure cloud computer, and one-click exports to spreadsheets, slides, PDFs, and images. Learn what changed, who gets it, and how small teams can use it.

Sham

Sham

AI Engineer & Founder, The Tech Archive

9 min read
0 views

Verdict: Google's June 2026 NotebookLM update turns it from a "chat with your PDFs" tool into an active research agent. It can now search the web for sources, run code in a secure cloud computer, and export charts, spreadsheets, slide decks, and images. It is live now for Google AI Ultra and qualifying Workspace business users; everyone else gets it later. For a small team, the most useful change is that you can hand it messy raw data and ask for a cleaned-up analysis, a chart, or a client-ready deck without touching another app.

Last verified: 2026-06-17

TL;DR

  • Engine: now runs on Gemini 3.5 and Google's Antigravity coding-agent framework.
  • New trick: every notebook gets a secure cloud computer that writes and runs code, with 100+ built-in software skills.
  • Output formats: PDF, Word, Excel, PowerPoint, Markdown, CSV, JSON, PNG/SVG charts, and Nano Banana images.
  • Source discovery: start with a blank notebook and a question; NotebookLM finds and cites web sources for you.
  • Availability: Google AI Ultra and Workspace business customers with AI Ultra / AI Expanded Access first; broader rollout planned.
  • Free tier: still exists at $0 (100 notebooks, 50 sources/notebook, 50 chats/day), but the new agentic features are paid-tier first.

What exactly changed in the June 2026 NotebookLM update?

Three things changed at once: the model, the tool's "hands," and how you start a project.

1. The brain switched to Gemini 3.5 and Antigravity

NotebookLM now runs on Gemini 3.5 and Google's Antigravity coding-agent framework. In Google's own side-by-side tests against the prior version, the upgraded system won more than 65% of the time across five core research tasks. It scored 69.9% on large-document analysis and 78.2% on advanced web research and source discovery. (Google internal evaluation, June 2026 — vendor claim.)

2. Every notebook got its own secure cloud computer

This is the biggest practical shift. Each notebook now has an isolated, sandboxed cloud computer. When you ask for a calculation, NotebookLM can write real code, run it, check the result, and hand back a chart or table instead of guessing. Google says the environment comes with 100+ curated software skills for tasks like parsing messy spreadsheets, comparing differently formatted documents, pulling numbers from PDFs, and building visual reports.

3. You no longer need to upload sources first

Previously, the tool only answered from files you supplied. Now you can open a blank notebook, type a research question, and NotebookLM uses Google Search to find relevant sources, add them, and cite them. You stay in control: it suggests sources, but you decide which ones go into the notebook.

Which output formats can NotebookLM now create?

The June update adds real file exports, not just text replies. Google lists the following formats in its announcement:

Output type Formats Best for
Documents PDF, DOCX, Markdown, plain text Reports, guides, briefs
Spreadsheets XLSX Budgets, models, cleaned data
Structured data CSV, JSON Feeding other tools / dashboards
Charts PNG, SVG Slide-ready visuals
Images PNG, JPG, GIF (via Nano Banana) Simple graphics from findings
Presentations PPTX Team decks

You can also edit the file after generation by giving follow-up instructions the same way you would to a colleague.

How much does NotebookLM cost in 2026?

NotebookLM is not sold on its own. It is bundled inside Google's AI plans. Google's official support table breaks the consumer tiers down like this:

Plan Price (USD/month) Notebooks Sources/notebook Chats/day Audio Overviews/day Deep Research
Standard Free 100 50 50 3 10/month
Plus (Google AI Plus) $7.99 200 100 200 6 3/day
Pro (Google AI Pro) $19.99 500 300 500 20 20/day
Ultra 20 TB (Google AI Ultra) $99.99 500 500 2,500 100 75/day
Ultra 30 TB (Google AI Ultra) $200.00 500 600 5,000 200 200/day

Pricing and limits change often. We last checked them on 2026-06-17 against Google's own help page. If you are reading this weeks later, verify the current numbers before choosing a tier.

The new agentic features (secure cloud computer, web source discovery, expanded exports) rolled out first to Google AI Ultra subscribers and Workspace business customers with AI Ultra Access or AI Expanded Access, according to the Google announcement.

Is the NotebookLM update worth paying for?

For most solo users and students, the free tier still covers note-taking and basic Q&A. The paid tiers start to make sense when one of these is true:

  • You hit the 50-source or 50-chat daily cap regularly.
  • You want to turn messy spreadsheets into clean charts or reports without touching Excel or Python.
  • You run market or competitor research and need NotebookLM to find and cite sources for you.
  • You create a lot of Audio Overviews, Video Overviews, or slide decks.

The $19.99/month Pro tier is the practical sweet spot for a small-business power user: 500 notebooks, 300 sources each, 500 chats/day, and 20 Deep Research runs/day. The Ultra tiers are enterprise-scale products; most small teams do not need them.

What this means for you

If you run a small business, side project, or lean team, the new NotebookLM is best used as a research and reporting assistant, not a replacement for your thinking.

Three high-value first moves:

  1. Clean up your messy data. Upload sales, refund, or lead-source spreadsheets and ask for the one metric that actually matters — e.g., "Which product is worth pushing harder?" The secure cloud computer runs the math instead of hallucinating a trend.
  2. Turn dull docs into usable guides. Feed it a long policy manual, onboarding doc, or technical spec and ask for a one-page team guide plus a slide deck. You get client-ready files without re-writing everything.
  3. Start research from a question, not a folder. Open a blank notebook, ask about a market or competitor, and let it find and cite credible sources. Then you edit the source list before trusting the answer.

The tool's real edge is speed: one person doing five jobs can hand off the grunt-work of formatting, calculating, and sourcing to the agent and spend their time on the decision.

How to use the new NotebookLM: a 5-step workflow

  1. Start with the outcome. Decide whether you want a report, chart, slide deck, spreadsheet, or cleaned data file before you type the prompt.
  2. Give it sources or a question. Either upload your files or ask it to find web sources on a topic. Review the sources it picks.
  3. Ask for a specific artifact. Use a prompt like: "Build an Excel file showing monthly revenue by product, with a chart and a short written summary."
  4. Check the work. Open the exported file and spot-check a few numbers or citations. The tool runs code, but you are still the editor.
  5. Iterate in plain English. Ask for changes the same way you would ask a teammate: "Make the chart a bar graph," "Add a column for gross margin," or "Shorten the summary to three bullets."

FAQ

Q: What is NotebookLM's "secure cloud computer"?

A: It is an isolated, per-notebook sandbox where NotebookLM can write and run code to analyze your data. Google says it is locked to that notebook, so your files and queries are not mixed with other users' data while the code runs.

Q: Can I use the new NotebookLM features for free?

A: The free Standard tier still exists, but the June 2026 agentic features — secure code execution, web source discovery, and expanded exports — rolled out first to Google AI Ultra and qualifying Workspace business users. Broader availability is planned, but Google has not set a date.

Q: What files can NotebookLM export now?

A: PDF, DOCX, Markdown, plain text, XLSX, CSV, JSON, PNG, SVG, JPG, GIF, and PPTX. You can also generate images with Nano Banana inside the notebook.

Q: Does NotebookLM still only use sources I upload?

A: No. You can now start with a blank notebook and ask it to find web sources for you. It adds them with citations, and you can remove any source you do not trust.

Q: How is this different from ChatGPT or Claude with file uploads?

A: NotebookLM is built around source-grounded answers and citations, and now around code execution inside a secure sandbox. The main trade-off is that it is less of a general chatbot and more of a structured research workspace. For a deeper comparison of AI assistants for small-business work, see our best AI assistant guide.

Q: Is my data safe in NotebookLM?

A: For individual accounts, Google's help page says data is protected and is not used to train NotebookLM unless you provide feedback. For Workspace and Education users, uploads, queries, and responses are not reviewed by humans and are not used to train AI models. Always review your own organization's data policy before uploading sensitive customer or financial information.

What this means for different roles

  • Founders / operators: Use it to turn raw business data into a decision-ready report. The secure code execution means you can trust calculations more than a plain chatbot reply.
  • Marketers: Feed it a batch of customer reviews, support tickets, or competitor pages and ask for a content brief, FAQ, or image that summarizes the main finding.
  • Consultants / coaches: Drop in call transcripts or training notes and ask for a one-page client guide or onboarding deck. For a similar workflow with a different tool, see our Claude 5-level progression guide.
  • Solo builders: Pair it with other AI assistants. If you already use Gemini, read our Gemini 2026 guide to see where NotebookLM fits in the same Google stack.
Sources
Updates & Corrections
  • 2026-06-17 — Article published. Facts verified against Google's official announcement and help-center pricing table.
  • 2026-06-17 — Availability noted as Google AI Ultra / qualifying Workspace business users first; broader rollout planned.

Get the practical AI brief

Verified, no-hype AI tips you can actually use - in your inbox. Free.

No spam. We verify what we send. Unsubscribe anytime.

Discussion

0 comments