Short-term rental listing clip: 8 seconds of Wan 3.0 is $1.00 at 720p

An 8-second clip on wan-3.0 costs exactly $0.50 at 480p, $1.00 at 720p and $2.00 at 1080p on Sume. Eight seconds at the 720p rate of $0.125 per second lands on a whole dollar with nothing to round, which makes it an easy unit to budget a rental listing around.

One clip per room is the pattern: living area, kitchen, bedroom, view. Four rooms at 720p is $4.00.

Price table for one clip

Wan 3.0 lists at $0.05, $0.10 and $0.20 per second by resolution and Sume bills list times 1.25 (video router docs). Eight seconds is on the clean side of the rounding rule, so these are exact.

Use your own photos as first frames

Guests compare the video with the photos, so a generated room that does not match is worse than no video. Send each real photo as a frame_images entry with frame_type: first_frame and keep the motion to a slow push-in or a pan. If you have a second photo of the same room from the other side, wan-3.0 also supports a last frame, so the clip can travel between two real views (video generation docs).

• One real photo per clip, first frame only, for a gentle push-in.
• Two real photos, first and last frame, for a walk through a doorway.
• Never invent amenities in the prompt. Describe camera movement and light, not furniture.

Disclosure and honesty

A listing video that shows something the property does not have is a problem on any platform. Treat the generated clip as an animation of real photos, watch every result for added or changed objects, and drop any take that does.

Sume returns the billed amount in usage.cost on the poll response, so you can log what each room cost.

When to pick 1080p

Most listing pages show video small. Choose 480p for a draft, 720p for the published clip, and 1080p only for a hero clip on your own site. The jump from 720p to 1080p doubles the price for the same eight seconds.

Practical notes

Order the clips like a tour: entrance or living area first, then the kitchen, then the bedroom, then the view or outdoor space. Eight seconds each gives a 32-second walkthrough from four jobs, which you can join in an editor with a single music track.

If a room photo is dark, brighten it before you use it as a first frame. The model starts from your image as it is, and a clean, well-exposed start gives you a better clip for the same $1.00.