18-second AI video API: draft at Wan 480p, then render final

An 18 second Wan 3.0 clip on Sume bills $1.13 at 480p, $2.25 at 720p and $4.50 at 1080p. That gap is the argument for drafting low and rendering the approved version once.

Which models accept 18 seconds

Eighteen seconds leaves only Wan 3.0 and Seedance 2.5 as generators; Genjutsu and Recast follow their source length.

A cheap draft loop

Wan's three resolution tiers are priced at list $0.05, $0.10 and $0.20 per second (list 2026-08-24, from the Sume pricing tables), so 1080p costs four times as much as 480p for the same seconds. A 480p draft tells you whether the motion and framing work before you commit.

The reserve-and-refund rule helps in the other direction: if a draft fails, the held amount is released. Keep the same prompt and aspect ratio across draft and final so the comparison is fair.

What a 18 second request costs

Sume bills the provider list price times 1.25, rounded up to the next cent. The amount is reserved when you submit and refunded if the job fails, so a rejected or failed clip does not cost you. The /v1/videos/models descriptor lists the billable rate in pricing_skus, which is the field to trust over any number in a blog post.

Three drafts at 480p cost $3.39 in total, which is less than one 720p render at $2.25.

Check 18 seconds against the live catalog

Duration support lives in the catalog, not in this page. Each entry on GET /v1/videos/models carries supported_durations, which is every whole second from the model minimum to its maximum, so one membership test answers the question. This script lists every model that accepts 18 s, with its resolutions and billable rates.

Set SUME_API_KEY first. See the Video Generation docs for the request and polling lifecycle.

import os
import requests

N = 18
resp = requests.get(
    "https://api.sume.com/v1/videos/models",
    headers={"Authorization": f"Bearer {os.environ['SUME_API_KEY']}"},
    timeout=30,
)
resp.raise_for_status()
for m in resp.json()["data"]:
    if N in m["supported_durations"]:
        print(m["id"], m["supported_resolutions"], m["pricing_skus"])