20 invented presenters for a persona test: $74.20 at Standard

Testing 20 different presenter personas on Sume costs $74.20 at Standard when each avatar is built from props and each test clip is 15 seconds: 20 x $0.95 = $19.00 for the avatars plus 20 x 15 x $0.184 = $55.20 for the videos.

What props means

Sume creates avatars three ways: a prompt, a Profile (the API's props input type, structured traits) or an image (photo). The docs example for props sets ethnicity, sex and age. No reference photo is involved, so there is no real person's likeness to clear. The result is a reusable avatar identified by avatar_handle.

Running the test

Create the 20 avatars as jobs, poll until each completes, then send each handle on a 15-second talking-video request with the same script, so the only variable is the presenter. Use one Idempotency-Key per avatar and per clip.

Each avatar creation is job-backed and returns a handle you choose; Sume normalizes it and stores it without a leading @. Use names that encode the persona, such as host_f28_asian, so results map back to the trait set.

• Same script, aspect ratio and scene for every candidate.
• Run Standard for the pass, re-render finalists at Plus or Max.
• Re-rendering 3 finalists at Max, 15 s each: 3 x 15 x $0.55 = $24.75.

After the test

The winner is already a reusable avatar. Subsequent videos only cost seconds, so the persona test is the one place where per-avatar $0.95 fees multiply. Budget it as a research line, separate from production video.

Practical notes

Idempotency. The docs examples send an Idempotency-Key header on every create call. Use one key per intended job and reuse it only when you retry the same request. A network timeout followed by a fresh submit with a new key is the one path that can create, and bill, a second job. Keep keys human-readable, such as the campaign and scene, so a support lookup is easy.

Quality tiers. quality accepts standard, plus and max. Plus is the default when you omit it. Standard is the fastest Sume execution path, and Max is the highest quality tier with slower turnaround. Because the three differ in both price and speed, decide the tier per stage of your workflow: a quick pass to confirm wording, then a final render once the script is settled.

Output format. Avatar video output is 720p at this time, and the aspect ratio defaults to 9:16 with 1:1, 3:4, 4:3 and 16:9 also accepted. Choose the ratio in the request rather than cropping afterward, since a crop throws away part of a frame you paid for by the second.

Job lifecycle. Every Sume avatar request is job-backed. You submit, store the job id, and poll GET /v1/jobs/{id}/status until it completes, then read GET /v1/jobs/{id}/result. The shared lifecycle also lets you wait with sync or subscribe mode for a limited time instead of polling, and the result carries Sume-hosted artifacts rather than provider URLs, so you never handle a provider queue id. Plan your client around the job, not around a single blocking response.