Recast 3 old 10-second clips with the face swap beta: cost by tier
Face swap beta reserves 15 seconds of the tier price per run, so 3 runs cost $8.28 standard, $11.025 plus, $24.75 max.
Recasting three existing 10-second clips with Sume's face swap beta reserves $8.28 at standard, $11.025 at plus and $24.75 at max. Each run reserves 15 seconds of the tier price, whatever the length of your clip. That is $2.76, $3.675 and $8.25 per run.
What the numbers are
The beta takes avatar_handle, video_url and quality, all required. The source clip should be about 4 to 15 seconds and have usable audio. Because the reservation covers a 15-second ceiling, a 10-second clip is reserved at 15 seconds and the catalog does not promise a lower final price.
| Tier | Per second | Per run (15 s) | 3 runs |
|---|---|---|---|
| standard | $0.184 | $2.76 | $8.28 |
| plus | $0.245 | $3.675 | $11.025 |
| max | $0.55 | $8.25 | $24.75 |
Is it cheaper to regenerate?
A new 10-second talking video at standard costs $1.84, against $2.76 for a swap run. If you have a script, regenerating is cheaper. Face swap is for when the original performance, framing and timing must stay and only the face changes.
- Use swap when the footage is already approved.
- Use a talking video when you can re-record the script.
- Test one clip at standard first; tiers differ in render quality, not in what the request accepts.
Notes on use
The beta route is POST /v1/models/sume/avatar-face-swap/v1.0/runs. Use an avatar you own or have permission to use, and follow the platform's disclosure rules for altered or synthetic content. YouTube describes its rules on the altered or synthetic content page. See the face swap docs.
Choosing a tier for recasts
Swap output is judged by whether the new face holds up against the existing footage. Start with the standard tier on the clip with the most head movement. If the face drifts, rerun that clip at plus and compare. The difference between standard and plus on a 15-second reservation is $0.915, which is a small cost for one test.
Max is $8.25 per run. It makes sense for a hero clip that will run as paid media, not for a back-catalog batch. For three clips, the spread between all-standard ($8.28) and all-max ($24.75) is $16.47.
Because audio matters for the source, check that each old clip has clear speech. A clip with only music is a poor fit for the beta, since the source needs usable audio.
Budget line for the batch
Put the swap in your budget as a flat per-run figure. At standard that is $2.76 per run, so ten clips would be $27.60 and thirty would be $82.80. The run is billed against a 15-second reservation, so the number does not move with the clip length inside the limit.
Sources
Related posts
More in Sume Avatar 1.0
- What $7.77 buys in AI avatar video on Sume: 42 seconds, or 14 at max
A $7.77 balance buys 42 s of standard avatar video, 31 s of plus or 14 s of max on Sume. Whole-second math for each tier, with and without a product image.
- Seven persona avatars, one 10-second clip each: $19.53 at standard
Seven reusable avatars at $0.95 each plus a 10-second video apiece: $19.53 at standard, $23.80 at plus on Sume. Handles and workflow.
- Three 15 s avatar hooks at standard ($8.28) vs one at max ($8.25)
On Sume Avatar 1.0, three 15 s hooks at standard cost 3 x 15 x $0.184 = $8.28, about the same as one 15 s render at max (15 x $0.55 = $8.25).
- AI doctor or expert on screen in a TikTok Shop health ad: banned
TikTok Shop's AI content page prohibits AI-fabricated expert or doctor identities endorsing health products. What to film, and what an avatar may still do.
Written by Sume