12 listing intros a month with an avatar: 20 seconds each costs $44.16

Twelve 20-second listing intro clips a month with Sume Avatar 1.0 cost $44.16 on the standard tier, $58.80 on plus and $132.00 on max, with no product image. Each clip is $3.68, $4.90 or $11.00. Add the property photos in a timeline render at $0.10 per started output minute if you want the avatar to introduce a photo tour.

These are list prices from Sume's avatar rate table (read 2026-10-06). The avatar clip itself is the intro; the listing photos are a separate step.

The monthly budget

Twenty seconds is about 56 words at the 2.8 words a second that Sume uses to estimate duration. That is enough for an address, three selling points and a call to action, which is all an intro needs.

Where the money goes wrong

Re-renders. If an agent changes a price or a bedroom count after the render, the whole clip is paid for again, because the words are baked into the audio and face. Lock the facts in the script before you spend, and use an Avatar video preview to approve the first frame at no render cost, then call generate-video from the preview id.

Defaults matter too. If you omit quality, the job runs on plus. A team that wants standard everywhere has to say so in every request, which is a good argument for one shared request builder.

A sensible workflow

Create the agent's avatar once and reuse its handle for every listing; creating the avatar is a separate one-time step that is not in the per-second table above. Render each intro on standard first, look at the first frame, and move only the flagship listings to plus. Then join the intro to the photo tour in Timeline 1.0, which accepts the avatar clip as a video slot because a finished avatar video is a media.sume.com artifact.

Two limits: the avatar renders at 720p, and one final video uses one avatar. If a brokerage wants two agents introduced together, that is two clips joined in a timeline, not one render.