The caller below
Photo 1 handles the challenge, the whisper and the upward call.
Two adults below trade a whisper and call upward while a separate villain reacts from the balcony.
Push the Button Rush Hour AI gives three adults distinct parts in the opening of the scene. The first photo becomes the floor-level caller holding the pistol, the second becomes the smaller adult beside them, and the third becomes the villain on the balcony.
The caller and villain use their own photo appearance and visible clothing. The second adult keeps their uploaded identity and hair within the scene's black uniform, white shirt, black tie and takeout-box vest.
All three photos are required, even though the characters appear in separate close-ups. The smaller role stays an adult at the scene's reduced scale; choose an adult portrait for that slot too.
The selected sequence begins with the caller's challenge, cuts to the smaller adult's reaction and the balcony villain's stare, then returns to the floor-level pair for the quick whispered exchange.
The caller turns upward again and the smaller adult joins in with arms spread. The window finishes during the return to the villain's close view, keeping the original cuts, speaking turns and reactions.
This version uses the first 15 seconds at the reference's pace. The later shouted punchline and the final side-by-side response belong to the longer scene and are outside this selection.
The grand hall, carved balcony, red panels, debris and scene props stay in place. Check which face appears in each cut, the whispering pair and the second role's vest before downloading with the reference dialogue.
Photo 1 handles the challenge, the whisper and the upward call.
Photo 2 takes the reactions and arms-out call in the fixed vest.
Photo 3 stays with the balcony close-up and listening expression.
Add one adult per photo in caller, uniformed adult and villain order.
Confirm all three people are 18 or older and agreed to appear, then generate for 10 credits.
Review the face order, whisper and balcony reaction, then download with sound.
Photo 1 is the caller with the pistol at floor level. Photo 2 is the smaller uniformed adult beside them. Photo 3 is the villain on the balcony. All three separate adult photos are required.
This version casts all three roles. Add one clear adult portrait in every slot before generating.
Use an adult photo for that role. The template keeps the character at the reference's smaller scale with the fixed uniform and takeout-box vest.
The caller and villain use their own visible photo clothing. The second adult uses their face and hair with the scene's black uniform, white shirt, black tie and connected takeout-box vest.
It uses the first 15 seconds: the challenge, reactions, whispered exchange, calls upward and return to the villain. The later shouted punchline and final paired response are outside this window.
The scene follows fixed speaking turns and reference dialogue in the grand hall and balcony. There are no script, voice, background or extra-person uploads.
A 15-second vertical 9:16 MP4 at 720p, at the reference's original pace, with audio from that same selected section.
One video uses 10 ReelTada credits for the three roles together. Your balance is checked before photo uploads and generation begin.
Three adult portraits, the whispered exchange and the opening standoff with sound.