Construction time-lapse from two site photos: Wan 3.0, $1.25
Use a start photo and a finished photo as first and last frame on Sume's wan-3.0. 10 seconds is $0.63 at 480p, $1.25 at 720p, $2.50 at 1080p.

A time-lapse between two real photos is a first-frame and last-frame job: send the empty lot as first_frame and the finished building as last_frame to wan-3.0. Ten seconds costs $0.63 at 480p, $1.25 at 720p and $2.50 at 1080p on Sume.
The model invents the in-between, so this is an illustration of progress and not a record of it. Label it that way if it will sit next to real project updates.
Request
frame_images carries both photos, and frame_type says which is which (video generation docs). Keep the camera position the same in both photos, or the model will have to fake a camera move as well as a build.
{
"model": "wan-3.0",
"prompt": "Time-lapse of construction, clouds moving fast, scaffolding rises, fixed camera",
"duration": 10,
"resolution": "720p",
"generate_audio": false,
"frame_images": [
{"type": "image_url", "image_url": {"url": "https://example.com/site-start.jpg"}, "frame_type": "first_frame"},
{"type": "image_url", "image_url": {"url": "https://example.com/site-done.jpg"}, "frame_type": "last_frame"}
]
}Price
Wan 3.0 lists at $0.05, $0.10 and $0.20 per second. Sume bills list times 1.25 and rounds up to the cent (video router docs).
| Resolution | Rate per second | 10-second clip | Five projects |
|---|---|---|---|
| 480p | $0.0625 | $0.63 | $3.15 |
| 720p | $0.125 | $1.25 | $6.25 |
| 1080p | $0.25 | $2.50 | $12.50 |
Tips for a believable result
Match the two photos in angle, time of day and framing. Ask for a fixed camera, and keep the prompt about the change: clouds, light, work crews as blurs. If the building appears to grow in the wrong order, shorten the prompt rather than adding detail.
Where this fits
It suits a marketing page or a pitch deck where the point is a sense of progress. It does not suit a progress report to a client or a lender. Use real photos on a schedule for those.
Practical notes
If the first run looks wrong, try a shorter clip. A 6-second time-lapse at 720p is $0.75 and gives the model less room to wander between the two photos.
Keep your originals. The two photos are the evidence; the clip is a presentation of them.
Sources
Related posts
More in Use cases
- Crop a 21:9 Seedance clip for TikTok: 1120x630 passes, 630x630 misses
A 720p 21:9 Seedance frame is 1470x630. Cropped to 16:9 it is 1120x630, above TikTok's 960x540 floor. Cropped to a square it is 630x630, ten pixels short.
- Customer onboarding email series with an avatar: 5 clips, cost by tier
Five 20-second onboarding clips from one Sume avatar cost $18.40 on standard, $24.50 on plus and $55.00 on max. How to plan the series and when to re-render.
- How do I automate a daily morning briefing audio with a TTS API?
A daily briefing of 2,200 characters is one TTS job: 11 cents a day, $3.30 for 30 days on Sume, plus a 1-cent concat if you add a music sting.
- DALL-E 2 was removed May 12: four variations from one edit call
OpenAI removed dall-e-2 on May 12, 2026. Replace it with one Sume edit call that returns up to four new takes of a photo, with estimated cost at each quality.
Written by Sume