The oh you can make it cap ai prompt is hard because it must keep two different people, many quick cuts and one perfectly timed spin consistent at the same time. On Oh You Can Make It Cap that work is already done inside each template. Make the video from one photo.
What does the prompt have to solve?
- Two identities. The scene cuts between two people. The model has to keep each one stable instead of blending them into a single face.
- The cuts. The clip alternates sides again and again in about 12 seconds. A single continuous shot will not do.
- The timing. The spin has to land when the other person reacts, and the straight-faced talking has to follow it.
- Your likeness. The person must still look like your photo from every angle, including the back of the head.
Why do copied prompts fail?
People post results with the two people merged, with a cut missing, or with the spin early or late. Each has a cause outside the words: the inputs that were supplied and settings that a pasted prompt never carries with it. Add the cost of retries, and a copied prompt becomes an expensive guess. If you still want to compare routes, the tool comparison lays them out.
What is built in here?
The motion, the shot list and the settings are fixed and tested in each template, so every render follows the same structure. Your only choices are the photo, the role and the format, as in the four steps. The prompt is not shown on this site, by design.