NotebookLM previously allowed users to generate Video Overviews from their uploaded sources, but the output was limited to narrated slides. Google has now expanded that capability significantly.
The new Cinematic Video Overviews generate fluid animations and rich, detailed visuals directly from your sources. The feature is powered by a combination of three AI models — Gemini 3, Nano Banana Pro, and Veo 3 — with Gemini acting as a creative director. It determines the narrative structure, visual style, and format, and refines its own output to maintain consistency throughout.
The update is particularly relevant for topics like history, geography, and biology, where static text often falls short. Users can now turn large volumes of source material into an immersive, fully animated video rather than reading through it manually.
The feature is currently available in English for Google AI Ultra subscribers who are 18 or older, across web and mobile platforms. Users can generate up to 20 cinematic overviews per day, and the NotebookLM app can be downloaded on iPhone and iPad via the App Store.