Skip to main content

Character plus set reference: walk to the window

Task
Reference to video
Resolution
1080p
Duration
8 s
Aspect ratio
16:9

Output

Input

  • Input image
    reference_image_urls
  • Input image
    reference_image_urls

Swap in your own when you run it.

PromptEN225 characters

The first image is the character — keep her face, hair and the black gown. The second image is the room. She walks from the doorway to the window and stops, looking out. Slow lateral track, one practical lamp, night ambience.

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

The prompt spends one clause on identity — 'keep her face, hair and the black gown' — and the rest on one action: walk from the doorway to the window and stop. `reference_image_urls` gives the woman first, the room second; `duration_seconds` `8` is entirely that walk and the slow lateral track after it. `seed` `640117` at `resolution` `1080p` keeps the room reproducible.

Settings used

SettingValueWhat it does
seed6401170 to 2147483647. Fix it while you tune resolution or duration, otherwise every rerun changes two things at once.
resolution1080p1080p, 2k or 4k, default 1080p. A native Wan 3.0 render tops out at 1080p, so the tiers above it — 2k and 4k — are delivered by enlarging that render instead of rendering natively at those sizes: the extra pixels come from the enlargement, and they are only as good as the render underneath. There is no 480p or 720p tier here; for those sizes, use Wan 3.0.
aspect_ratio16:916:9, 4:3, 1:1, 3:4, 9:16 or adaptive. Defaults to adaptive, which lets the model choose; name a ratio when the delivery format is fixed.
duration_seconds8Any whole number from 2 to 30, default 5, or -1 for smart duration, which reserves the maximum and settles on the length actually delivered. With reference videos, their total plus the output must stay within 30 seconds and both are billed. Find the shot at 5 seconds and 1080p before a long 4K render.