Face swap API: quality has no default, unlike avatar videos

On Sume's Beta face swap endpoint, quality is required and there is no default: send standard, plus or max. That differs from avatar videos and avatar video previews, where omitting quality falls back to plus. If you copy an avatar request body over to face swap, add the field.

Where is quality required and where does it default?

The face swap docs list avatar_handle, video_url and quality as required and state "There is no omit default on this endpoint." The avatar video docs say plus is the default when omitted.

What do the three values mean?

For avatar videos the docs describe standard as the fastest Sume execution path, plus as the balanced one and max as the highest quality tier with slower turnaround. The face swap page names the same three values but does not restate what each does, so this post only claims the avatar-video descriptions for avatar videos.

What does the request look like?

Set quality explicitly in the body of POST /v1/models/sume/avatar-face-swap/v1.0/runs.

{
  "avatar_handle": "reference_presenter",
  "video_url": "https://example.com/source-clip.mp4",
  "quality": "standard"
}

What happens if I leave it out?

The docs do not print the error text for a missing quality, so read the response body of a test request for the exact code rather than assuming one. Because the endpoint is Beta, check the page again before you depend on this behavior.