Swap a spokesperson in a 14-second clip: face swap beta $2.58 to $7.70

Swapping the spokesperson in a 14-second regional ad clip with Sume's Avatar Face Swap beta costs $2.58 at standard, $3.43 at plus and $7.70 at max, using $0.184, $0.245 and $0.55 per second. The source must be about 4 to 15 seconds with usable audio, and you need a ready Avatar 1.0 identity whose face you have the right to use.

What the endpoint takes

The endpoint is POST /v1/models/sume/avatar-face-swap/v1.0/runs. The required fields are avatar_handle, video_url and quality, and the beta has no default quality, so omitting it fails. The source must be a public HTTPS video URL; localhost, private networks, signed or private URLs and provider task URLs are rejected.

By design, the endpoint does not accept prompts, transcripts, duration knobs, aspect ratio or provider fields. It applies a face to a clip. If you need a new script spoken, that is the avatar talking-video endpoint instead.

Regional variants

For one ad in five markets with a different local spokesperson, each variant is its own job: five at standard is 5 times $2.576, which is $12.88. Draft at standard, look at the face match and the mouth, and re-run only the keepers at a higher tier.

Face swap changes the face, not the speech. If the language differs by market, the audio needs separate handling, so check that each market's source clip already carries the right voice.

Rights and disclosure

Use only avatars and source footage you have permission to use, and check the rules of the ad platform and the market for synthetic or altered media before you run a campaign. This post covers the mechanics and price, not clearance. Poll GET /v1/jobs/:id/status, and read the result at /result when ready.

A practical note

Source quality drives results more than the tier. Use a clip with a clear, front-facing subject, steady light and clean audio, and keep it inside 4 to 15 seconds. A face that turns away or is covered by a hand is harder to replace. Preview a single market at standard first. If the result is acceptable, the plus and max tiers show whether the extra $0.86 or $5.12 for 14 seconds buys a visible difference on your footage. Poll with the job status endpoint and keep the receipt for each variant for your own records.

Before you scale

Every price in this post is a Sume list price read from the public catalog on 2026-10-09, and the arithmetic is shown so you can redo it with your own counts. Rates can change, so re-read the catalog before a large batch and run a small test first. Sume bills the job's captured amount, and the job result tells you what was used, so compare the first run's receipt with your estimate before you scale the work to the full set.