ByThen

ByThen is built for creators who publish narrator-led video at volume: faceless YouTube channels, explainer and documentary content, animated storytelling, and comedic sketch formats. Instead of stitching together several separate AI tools, ByThen runs a five-step workflow inside one subscription. You start with a creator setup that locks your niche, language, format, art style and voice, then move through script and voice selection, key asset generation, storyboard review, and final video assembly.

Consistency is the platform's central claim. Characters, colour palettes and art style are fixed at the beginning and held across every scene, so a character does not drift between clips. Visuals are generated from what the narration actually says, meaning the on-screen content matches the spoken script. ByThen also keeps a saved creator profile of characters, environments and assets, so new projects start with everything already in place.

Output runs from 30 seconds to 30 minutes per generated video, with projects editable and exportable up to 60 minutes at a locked 25 FPS and 720p resolution, with 1080p and 4K listed as coming soon. There is a style library of over 100 art styles plus custom references. A built-in multi-track editor lets you trim, rearrange, overlay your own footage, adjust captions and add sound effects without regenerating the whole video. ByThen also exposes a native Model Context Protocol server so AI agents such as Claude can generate, edit and orchestrate projects programmatically.