Last frame to first frame: how AI video continuation works
Why the frame matters more than the prompt
Two runs from the same prompt will not match. Wording describes a scene; it does not pin down where the camera was, how the light fell or where the subject stood at the final moment.
A frame pins all of it. Handing the previous take's last frame to the next run as its first frame means the new shot inherits the composition exactly, and the prompt is left to describe what should happen next rather than what things look like.
What you see in the interface
The chain labels each join so the relationship is explicit: this take's last frame seeds the next run. The two takes below are a real pair, the second seeded from the final frame of the first.
Frequently Asked Questions
Can I pick a different frame to continue from?
The editor captures a frame at the playhead as well, so you can grab a specific moment and use that as a starting image. The Extend button is the one-press version that always takes the final frame.
Does the next run need a prompt too?
Yes, and that is the point. The frame fixes what the scene looks like; the prompt says what happens next.
Your first clip is the hard part
Open the Video Studio, get one shot you like, then press AetherWave Extend and keep going. Every take is its own gallery entry, so nothing you keep can be overwritten.
Open Video Studio