Gemini Notebook Can Now Build Interactive Websites from Your PDFs and Notes
Gemini Notebook, formerly NotebookLM, can now generate interactive websites from your materials: PDFs, notes, and links. Type "make me a site" in the chat and describe the result you want, such as a statistics dashboard or a visual study guide. The page opens in a separate tab, and you can share it by link. For students and researchers, this is a way to turn a stack of documents into visual material without writing any code.
AI-processed from TG: @notboring_tech (Not Boring Tech); edited by Hamidun News
You have a folder of PDF articles, study notes, and bookmarks, and you need to quickly show colleagues, classmates, or a client a clear picture of it all. Usually you would have to build a presentation by hand, draw charts, and retell the documents in your own words. Gemini Notebook, formerly NotebookLM, aims to shorten that process to a single phrase in the chat.
What Has Been Added
According to a post from the Telegram channel Not Boring Tech, Gemini Notebook can now turn uploaded sources into interactive websites. The examples include statistics dashboards and visual study guides, but judging by the wording, the list is not limited to these two formats.
The workflow is simple. You upload your materials to the notebook: PDFs, notes, and links to web pages. Then you type "make me a site" in the chat and describe what you want. Say you have ten sales reports and need a dashboard with the key figures. Or you are preparing for an exam and want a study guide where topics expand on click instead of scrolling through long text. The finished page opens in a separate tab and can be sent to another person by link.
The value lies in the fact that the answer is built from your own documents. The notebook works with what you put into it, so the result is tied to specific sources rather than the model's general knowledge. For studying and research, this is more convenient than asking a chatbot to "make a site about the topic" and then checking the facts by hand.
Why This Is Possible Only Now
The post's author says the developers could not reach high quality for this feature on the Gemini 3.8 model for months. With the release of Gemini 4 Hydrogen, this reportedly finally worked. The post gives no details: it is unclear what exactly was failing on the previous model: layout, numerical accuracy, or the stability of interactive elements. So this should be read as the product team's own account, not as an independent comparison.
The underlying idea is clear. To produce a working interactive page on request, the model must simultaneously understand the content of the documents, choose a suitable form of presentation, and output code that does not break when opened. A mistake at any step turns a "smart site" into a broken picture. If the developers were really struggling with quality, releasing the feature only after switching models looks logical.
What It Means in Practice
For students and teachers, this is a quick way to turn lectures and articles into material that is convenient for reviewing a topic. For analysts and product managers, it offers the ability to build a simple dashboard or overview page from a set of documents without involving a developer or designer. For small teams, it is a way to share research results with a single link instead of email attachments.
There are also questions the source does not answer. It does not say which plans or countries the feature is available in. It is unclear whether the generated site can be edited manually or its code exported, who can open the link, and how this fits with the privacy of uploaded documents. The last point matters especially: if the notebook contains internal reports, check the access settings before sharing the page.
Another piece of advice is not to trust the figures on a dashboard blindly. Any generation on top of documents can make mistakes when extracting data, so key values are better verified against the original PDFs. Individual numbers and labels take a couple of minutes to check, while an error in an "official-looking" dashboard costs more.
In short, Gemini Notebook takes one more step from a chatbot that answers in text toward a tool that produces a finished interactive result. How well this works on real, imperfect documents will become clear from the first user reviews. You can try it on the service's page: notebook.google.com.
- Nov 21, 2026California Sues OpenAI: Its AI Agents Escaped Control at Least 80 Times
- Nov 21, 2026MIRA Neural Network Renders Rocket League at 20 FPS, but Ping Hits 460 ms
- Nov 21, 2026Google built a closed copy of the internet with fake banks to train Chrome's AI agents
- Nov 21, 2026MLPerf Training Adds Nvidia Rubin and TPU Ironwood in First Appearance: 153 Systems
Key Questions Answered
What Has Been Added?
According to a post from the Telegram channel Not Boring Tech, Gemini Notebook can now turn uploaded sources into interactive websites. The examples include statistics dashboards and visual study guides, but judging by the wording, the list is not limited to these two formats. The workflow is simple. You upload your materials to the notebook: PDFs, notes, and links to web pages. Then you type "make me a site" in the chat and describe what you want.
Why This Is Possible Only Now?
The post's author says the developers could not reach high quality for this feature on the Gemini 3.8 model for months. With the release of Gemini 4 Hydrogen, this reportedly finally worked. The post gives no details: it is unclear what exactly was failing on the previous model: layout, numerical accuracy, or the stability of interactive elements. So this should be read as the product team's own account, not as an independent comparison. The underlying idea is clear.
What It Means in Practice?
For students and teachers, this is a quick way to turn lectures and articles into material that is convenient for reviewing a topic. For analysts and product managers, it offers the ability to build a simple dashboard or overview page from a set of documents without involving a developer or designer. For small teams, it is a way to share research results with a single link instead of email attachments. There are also questions the source does not answer.
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.