Listing clip from photos with Seedance 2.5, and what to disclose
Nine listing photos can feed a Seedance 2.5 reference-to-video request on Sume. Price a 30 s clip, plan the camera path, label it AI-generated.

You can turn up to nine listing photos into a 30-second walkthrough-style clip with Seedance 2.5 on Sume by sending them as reference_image_urls and describing a slow camera path. At 720p that is about $17.33 per clip on Sume's rate card. The clip is a generated interpretation, not a tour of the real rooms, so label it as AI-generated and keep the original photos next to it.
What the model can and cannot do here
BytePlus lists e-commerce, advertising and previsualization among Seedance 2.5 use cases (BytePlus: Dreamina Seedance 2.5). A reference-to-video model uses your photos as guidance; it is not a photogrammetry tool. It may rearrange furniture, change window positions or invent a hallway that does not exist.
For real estate that is a real risk, because a buyer may treat motion as proof of layout. Do not use a generated clip to claim square footage, room adjacency or condition.
Plan the shot list
Choose photos that show different rooms from similar eye height, keep the same daylight, and remove clutter before upload. Ask for one simple camera move per beat, such as a slow push-in or a gentle pan. The Video Router docs list up to 10 reference_image_urls and up to 3 reference_video_urls of at most 3 seconds each for reference-to-video, so nine photos fit; read the model's row in the catalog for its exact limits.
- Photo 1: front exterior, used for the opening.
- Photos 2 to 4: living area, kitchen, dining.
- Photos 5 to 7: bedrooms and bathroom.
- Photos 8 and 9: outdoor space and a detail shot.
The request and the price
A draft at 480p costs about $8.06 for 30 seconds; the 720p version about $17.33; 1080p about $42.65 (all computed from Sume's token pricing at list x 1.25). Draft first, then pay for the resolution that the listing platform actually shows.
| Resolution | 15 s | 30 s | 20 listings at 30 s |
|---|---|---|---|
| 480p | $4.03 | $8.06 | $161.21 |
| 720p | $8.67 | $17.33 | $346.68 |
| 1080p | $21.32 | $42.65 | $852.93 |
curl -X POST https://api.sume.com/v1/video-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: listing-clip-001" \
-d '{
"model": "seedance-2.5",
"prompt": "Slow walkthrough: front exterior, then the living area and kitchen, soft daylight, gentle camera glide.",
"reference_image_urls": ["https://example.com/ext.jpg","https://example.com/living.jpg","https://example.com/kitchen.jpg"],
"resolution": "720p",
"duration": 30,
"aspect_ratio": "16:9",
"mode": "async"
}'Disclose and keep proof
Add an on-screen note such as "AI-generated visualization" with authored caption cues, since a silent clip cannot be captioned from speech. Keep the source photos and the job id so you can show how the clip was made. Check your local listing rules; this post is not legal advice.
Sources
Related posts
More in Use cases
- Real estate price-reduction clip from one listing photo: $0.83 each
A 5-second vertical price-drop clip from the hero photo with the new price burned in as cues: $0.825 per listing, $9.90 for 12, on Sume from docs.
- Rehearsal dinner slideshow: 24 photos and Timeline's 8-fade cap
A 24-photo slideshow with fades fails if every cut is a fade: Timeline refuses more than 8 chained transitions. Insert hard cuts; 96 seconds costs $0.20.
- 40 fixed phone prompts in one TTS job: no streaming needed
Fixed prompts do not need 150 ms streaming. Render them once as sentence slices with Sume TTS and play files. Worked cost for 40 prompts of 60 characters.
- Re-render only the SKUs that changed: hash rows for a Sume bulk run
Retail surveys cite inventory imbalance and reactive decisions. Hash each product row, then send only changed rows to a Sume bulk queue with a revision key.
Written by Sume