New reframe engine, preview matches export exactly
A rebuilt reframe engine renders your clip identically in the live browser preview and in the final export, what you see is exactly what you get, with no surprises after rendering.
Katto turns long videos into scored, captioned vertical clips. Here's what we've already shipped and what we're building next, from AI captions and smart reframe to dubbing, 4K export and a public API.
Vote on what's next →A rebuilt reframe engine renders your clip identically in the live browser preview and in the final export, what you see is exactly what you get, with no surprises after rendering.
Give each scene its own framing, full-screen face tracking, wide, split-screen, stacked gameplay + webcam, or a 3- or 4-person grid, instead of one layout for the whole clip.
Click the video and drag a crop box with corner handles to pan, zoom and resize the framing of any scene, with the source shown around the frame for context.
Position each speaker in split-screen, place the gameplay and webcam in stacked gaming layouts, or arrange up to four people in a grid, all by dragging each region.
A cleaner, faster editor with a distinct video panel, dark timeline lanes, a drag zoom slider and an Apply-changes flow, polished to feel like a real editing tool.
Fine-tune how the auto-reframe camera follows your subject with follow-speed and dead-zone sliders.
Tick the passages you highlighted, choose a ratio, and Katto cuts them into a clip of their own without going back to the source video.
Pull another moment of the source video into a clip you are already editing, instead of starting a new one.
Choose split, grid or stacked from small previews built out of the shot you are on, instead of reading a list of names.
A crop you draw by hand, and the panes, cells and bands you drag, stay attached to the shot you set them on and survive a reload.
A landscape image or video used as b-roll is filled with a blurred background instead of hard black bars.
Remove a file you uploaded from storage, not just from the list in the editor.
Every shot of the clip, its time range and its framing, in the side panel rather than a menu over the preview.
Expand AI dubbing far beyond the current 19 languages.
Save your captions, logo, intro and colours once and auto-apply them to every future clip.
A visible safe zone in the preview so captions and faces never get cropped on any platform.
Word-level captions, B-roll and audio on real, editable timeline tracks.
Clean fades, dissolves and zoom transitions between scenes.
Generate click-worthy YouTube and short-form thumbnails in seconds.
AI upscaling for crisp 4K clip exports.
Select a time window from a multi-hour Twitch or Zoom recording before clipping.
Upload and use your own brand typefaces in captions.
Seats, shared projects and a shared brand kit for teams and agencies.
See which clips actually landed with cross-platform performance analytics.