Face swap API has no prompt or aspect_ratio: the three-field body

The Avatar Face Swap beta takes only avatar_handle, video_url and quality. By design it does not accept a prompt, transcript, duration, aspect ratio, avatar id or provider field, and quality has no default. Plan on reserving the 15-second price, which is $2.76 on standard.

The request

POST /v1/models/sume/avatar-face-swap/v1.0/runs is the only path; it is not the old consumer /face-swap route. The source video_url must be a fetchable public HTTPS video of about 4 to 15 seconds with usable audio. Localhost, private-network, signed or private URLs and provider task URLs are refused.

curl -X POST https://api.sume.com/v1/models/sume/avatar-face-swap/v1.0/runs \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: face-swap-001" \
  -d '{
    "avatar_handle": "product_host",
    "video_url": "https://example.com/clip.mp4",
    "quality": "standard"
  }'

What it reserves

The catalog prices a face swap at the Avatar Video no-product rate times the 15-second beta maximum, whatever the length of your clip. Avatar rates are Sume's charged per-second tier rates from the repo rate card (standard $0.184, plus $0.245, max $0.55 without a product image); confirm them in GET /v1/catalog.

Limits and gotchas

• If you need a scripted presenter instead of a recorded one, use the talking video route.
• To change the language or the words, record a new source clip; this endpoint cannot rewrite speech.
• Only use faces and footage you have the right to use.
• Poll GET /v1/jobs/:id/status, then read video_url from the result; use resource_status as the readiness signal.