21-second AI video API: Seedance 2.5 or Wan 3.0 by aspect ratio

A 21 second clip on Sume comes down to Seedance 2.5 or Wan 3.0, and the deciding field is aspect ratio. Seedance 2.5 lists eight values: auto, adaptive, 21:9, 16:9, 4:3, 1:1, 3:4 and 9:16. Wan 3.0 lists the same set without 21:9.

Which models accept 21 seconds

Genjutsu and Recast follow their source clip, so they only reach 21 s if the source is 21 s long. For a plain prompt-to-video generation, only the two rows below qualify.

Pick by frame shape

If you want ultrawide cinematic output, Seedance 2.5 is the only generator here that offers 21:9. For vertical, square or classic 16:9 and 4:3 work, both models qualify and the price decides.

The /v1/videos/models descriptor filters ratios to an OpenRouter-compatible list, so auto and adaptive do not appear there even though the Video Router catalog lists them. The stored post on Seedance 2.5 ratios explains the difference.

What a 21 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.

Twenty-one seconds of Wan bills $1.32 at 480p, $2.63 at 720p and $5.25 at 1080p. For Seedance 2.5 read pricing_skus.

Check 21 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 21 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 = 21
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"])