Skip to main content

1920s casino: two characters at roulette from three images

Model
Wan 3.0
Task
Reference to video
Resolution
720p
Duration
5 s
Aspect ratio
16:9

Output

In a gilded 1920s casino under chandeliers, a woman in a deep-V fringed black dress leans on a man in a tailcoat at the roulette table, blows on the ball and laughs as the wheel spins

This preview is blurred

This preview is marked sensitive. Confirm your age to see it.

Input

  • Input image
    reference_image_urls
  • Input image
    reference_image_urls
  • Input image
    reference_image_urls

Swap in your own when you run it.

PromptEN515 characters

Image 1 is the woman and Image 2 is the man — keep both faces and their clothes exactly. Image 3 is the casino. At a roulette table in Image 3 she leans against his side and bends close to blow softly on the ivory ball between his fingers; he spins the wheel and flicks the ball in, she tips her head back and laughs, and her fingers slide slowly down his loosened bow tie. Slow push in, gold chandelier light, the crowd cheering around them. Audio: a jazz big band, the roulette wheel rattling, the crowd cheering.

Written in English, exactly as we sent it.

Run this example

Checking the price…

Opens the playground with every setting filled in and shows the exact price before you run.

Same model, starting from this prompt. Other settings use their defaults.

Why this prompt works

Three references, three roles, named in upload order: Image 1 is the woman, Image 2 the man, Image 3 the casino. Her finger waves and fringed black dress, his tailcoat and loosened bow tie, and the gilded dome, chandeliers and roulette tables all come from the pictures; the prompt adds only the blocking — she blows on the ball for luck, he spins the wheel, she laughs and runs her fingers down his bow tie. `duration_seconds` `5` at `720p`, `16:9`.

Settings used

SettingValueWhat it does
resolution720p480p, 720p or 1080p. Defaults to 720p. Resolution changes both the price and how long the task takes to come back.
aspect_ratio16:916:9, 9:16, 1:1, 4:3, 3:4 or adaptive, which lets the model choose. Defaults to 16:9.
generate_audiotrueOn by default. The clip comes back with sound already locked to the picture. Set false when you are scoring the shot yourself.
duration_seconds5Any whole number from 2 to 30, or -1 for smart duration. Defaults to 5. With reference videos, their total length plus this value must stay within 30 seconds, and those seconds are billed on top of the output. Settle the look at 5 seconds before extending.