AI video generator for real estate: which Sume Format?
Sume's 27 catalog Formats include none for property listings. What to call instead for a listing video, and the disclosure test agents face.

There is no real estate Format in Sume's catalog. The ready-made Formats at the reserved sume handle are 27 slugs, and none of them is a listing tour, an open-house reel or an agent intro. If you search for an AI video generator for real estate, the honest Sume answer is: use the video API on your listing photos, or fork a catalog Format into your own and call it at {your_handle}/{slug}.
This post reads the catalog on 2026-10-06 and shows how to check it yourself, so you do not rely on a slug that does not exist.
What is in the catalog, and what fits a listing?
The Format catalog says any slug not on its list answers 404 format_not_found at sume/{slug}. The list is product ads, beauty stills, try-on and fitting videos, logo motion, slideshow, wall-of-text, green-screen and video-hook Formats, a fruit-character clip, and two video-to-video Formats (Recreate and Restyle). The descriptions below paraphrase the card text.
| Slug | What the card describes | Fit for a listing |
|---|---|---|
sume-product-commercial | Premium product commercial: hero composition, one cinematic camera move | Closest in style, but built for a product, not a room |
sume-cinematic-studio-commercial | Premium light, macro detail, high-end camera move | Product and luxury ads; not interiors |
sume-logo-motion-design | Animates a supplied logo into a short ident | Useful for a brokerage end card |
sume-restyle | Restyles a clip you own: same motion and cuts, new look | Re-grade your own walkthrough footage |
sume-recreate | Rebuilds a reference video with new people, voice and music | Not for property footage |
What do you call instead?
For stills-to-motion, POST /v1/videos takes frame_images for first and last frames, and GET /v1/videos/models lists which models accept them. A room-to-room move is a first-frame and last-frame job. Read the catalog before you submit, because durations and ratios differ per model.
If you want a repeatable recipe with a typed receipt, author a Format of your own. The catalog page says to change a catalog Format you fork it in the Format library, and your copy is addressed {your_handle}/{slug}. You then call it like any Format: one POST with instruction, input, an optional output_schema and a spend cap.
curl -sS "https://api.sume.com/v1/formats" \
-H "Authorization: Bearer $SUME_API_KEY" \
| jq -r '.data[] | [.handle, .slug, .io.profile] | @tsv'What about disclosure?
HousingWire's 2026-06-30 piece says AI-generated or AI-animated listing video, such as a drone-style move built from still photos, needs the same care as an edited photo. It proposes a five-part test, including whether the buyer will actually see the disclosure and whether the agent can show what was real and what was changed, and cites California AB 723, effective 2026-01-01.
Nothing in Sume adds that disclosure for you. Put the line in your caption or in the output your Format returns, and keep the unaltered photos. Sume's receipts record what ran, not what the law asks you to say.
Where does this leave you?
Pick the video API if you have photos today and want one clip per listing. Pick your own Format if you will run this weekly and want the same recipe, a typed result and a cap on spend. Do not call a product Format and expect it to know floor plans. Test one listing first and look at the frames.
Sources
Related posts
More in Use cases
- Real estate listing video with an AI avatar: fair housing script check
Describe the home, not the buyer. A fair housing word check in Python, a listing-photo scene on Sume Avatar 1.0, and the cost of a 45-second listing clip.
- Reddit 15-second ad: join hook, demo and offer, then plan first
Build a 15-second ad from three clips on Timeline 1.0 and check its length and cost with the unbilled /plan call before you pay for a render.
- Reddit 15-second engaged views: does your voiceover CTA land?
Reddit's Engaged Video Views beta bills videos over 15 s at 15 s. Use TTS word timestamps to check that your call to action is spoken before second 15.
- Reddit 15-second video ad: burn captions so it reads on mute
For a 15-second Reddit video ad, burn captions into the cut with POST /v1/video-captions: one $0.20 job, style slam for English, no new edit.
Written by Sume