Why it helpsNot every topic wants the same kind of video
Updated November 2026
Upload up to 10 PDFs or notes (20 on Laureate) and choose the Freestyle video mode. Scholarly reads the source and selects the visual style and animation approach that fits it, such as kinetic type, a whiteboard, a clean diagram sequence, or a blend. It then renders a narrated, chaptered video with transcript, captions, checkpoints, AI chat, and flashcards. A 2-minute Freestyle video is usually ready in about 6 to 11 minutes. The Free plan includes one lifetime AI Creation Credit at 720p with a roughly 2-minute length; paid plans add 1080p, watermark removal, MP4 download, and a target length up to 45 minutes.
A glossary of legal terms, a four-stage manufacturing process, and a chapter on cognitive biases are not the same kind of material, and a single fixed visual treatment serves at most one of them well. Freestyle starts from the content and works outward to the look, rather than the other way around.
That freedom applies to style only. Every Freestyle lecture still follows one coherent, linear teaching arc: it sets up the topic, builds the ideas in order, and closes with what you should now be able to do. The narration is grounded in your source, so what it says matches what your notes say.
If a lecture does not land the way you hoped, switching modes and regenerating is often a bigger improvement than rewording the prompt. Freestyle is the natural place to try when the named modes have not fit - and because the same study tools come with every mode, nothing is lost by experimenting.
What a Freestyle explainer includes, and what it leaves out
An explainer video takes one idea, document, or process and makes it clear to someone new to it. Scholarly makes explainers from your own material in any of its styles. Freestyle is the one where the AI also chooses how the explainer looks.
- What goes in: a PDF, notes, a deck, a link, a YouTube video, or a typed topic. Free uses one source per video; Premium combines up to 10 and Laureate up to 20.
- What comes out: a narrated explainer with chapters, captions, a transcript, and question checkpoints. Free videos run about 2 minutes; paid plans set a target up to 45 minutes.
- No presenter: an AI voice narrates over generated scenes. Nobody appears on camera, and there is no avatar.
- Not a product-demo or ad maker: there are no screen recordings, brand templates, or stock footage. It is built to explain a concept, a document, or a process.
- Editing: there is no timeline editor. While the video is being made you can ask for a change, and afterwards you can remake it in another style, voice, or language as a new version.
Pick by the job: the AI lecture generator for a full lesson, the short video generator for a 30 to 45 second clip, the AI training video generator for work documents, or essay to video for an essay or paper.
For work and teams
Product explainers, internal training, and process walkthroughs rarely fit one house style - a pricing change, a new support workflow, and a security refresher each want a different treatment. Freestyle chooses it per document. Pair it with the training video generator for a structured curriculum, or see Scholarly for Teams for shared source libraries, roles, and one bill.
