Cameo in an AI episode: Avatar Face Swap (Beta), 4 to 15 s source
Sume's Avatar Face Swap (Beta) puts your avatar into a source video of 4 to 15 seconds and requires a quality setting. Use it for a cameo, not a whole episode.
Sume's Avatar Face Swap (Beta) applies a ready avatar's face onto a public source video of roughly 4 to 15 seconds, and it requires a quality field. That makes it a tool for a cameo, such as the series presenter appearing in one shot you already have, and not for a whole episode. It runs at /v1/models/sume/avatar-face-swap/v1.0/runs.
The facts
The request takes avatar_handle, video_url and quality (standard, plus or max). There is no default for quality; leave it out and the call is refused. The endpoint has no prompt, transcript, duration or aspect ratio field, so what you send is the source video as it is.
| Item | Detail |
|---|---|
| Route | /v1/models/sume/avatar-face-swap/v1.0/runs |
| Status | Beta |
| Source video | Public HTTPS, approximately 4 to 15 seconds with usable audio |
quality | Required: standard, plus or max; no default |
| Avatar | A ready Avatar 1.0 avatar, passed as avatar_handle (creation is $0.95, once) |
Where it fits
A series that has a presenter can use the swap to place that presenter into a shot you already have: an establishing clip, a reaction or a stock scene. Keep each cameo to one shot. For a longer source, split it into segments of 15 seconds or less, run each, and join them in a timeline.
Cameo or full recast
If you want a different person to replace the lead across a finished episode, that is a different tool: the H3 Max Recast model on /v1/videos, which takes one to four person photos. Face Swap is for your own avatar. Pick by the question: 'put my presenter in' is Face Swap, 'replace the actor with this person' is Recast.
The source must be a public HTTPS video URL. Localhost, private-network, non-HTTPS and signed or private links are rejected, so host the clip somewhere durable first. The docs say the current plan is roughly 4 to 15 seconds of source with usable audio, so a silent clip is a poor fit.
Poll the job at /v1/jobs/{id}/status and read the result when resource_status says it is ready; the finished video lives under media.sume.com.
Before you use it
- Beta means the behaviour and fields can change; read the face swap page of the docs before you build against it.
- Use only faces you have the right to use. A cameo of a real person needs their consent.
- Pick a source with clear audio and a visible face, since the docs describe the source as one with usable audio.
- Set the quality value in your season config, so the choice is made once.
Sources
Related posts
More in Sume Avatar 1.0
- Griffin-style follow-ups with rendered avatar clips and branching
Griffin-Lite reacts live. Until you can use it, approximate a guided conversation with a set of pre-rendered Sume avatar clips and your own branching logic.
- Interactive avatar or avatar video? A five-question test
Griffin-Lite is a research preview. Five questions tell you whether you need a live avatar or a rendered Sume Avatar 1.0 clip, and what to ship this quarter.
- Name your avatars: a handle scheme that fits 2 to 30 characters
Sume avatar handles allow letters, digits, periods and underscores, 2 to 30 characters. Build a team naming scheme that passes validation and stays readable.
- Let users pick an avatar in your app with GET /v1/avatar-1.0/avatars
Build an avatar picker on the Sume Avatar 1.0 list route: read ready avatars server-side, cache them, and pass the chosen handle to talking-video.
Written by Sume