An extra person in the shot, a caption, a logo, a cable or clutter on a desk: just write what to remove. The AI erases the object across the whole clip and fills in the background, with no frame-by-frame masking.
The object is described in words. You do not have to outline it on every frame as in a regular video editor.
Where the object was, the AI draws a plausible background that carries the scene on.
The same editor can extend a clip, change the frame format, add subtitles or sound.
People, objects, captions and other things you can describe in words. The more precise the description, the cleaner the result.
The model is meant to leave everything else as it is, but complex scenes can show small differences, so compare the result with the original.
No, you pay for each run from your balance.