7-second AI video API: which Sume ids take 7 s and the price

Seven seconds is a valid clip length for all twelve public Sume ids, including Gemini Omni Flash 1.1, MiniMax H3 and H3 Max, Wan 3.0, Kling 3 and the Seedance 2 family. It is not valid on Google's Veo 3.1, whose Gemini API page lists durations of 4, 6 or 8 seconds (read 2026-10-04).

If your storyboard says 7 s because a voiceover runs 6.8 s, you do not need to round to Veo's grid. Pick a model whose range includes 7 and send the exact value.

Which models accept 7 seconds

The table lists the models that accept 7 s, with what a single 7 s clip bills. Gemini Omni Flash 1.1 takes 3-10 s, so 7 s is mid-range; the H3 family starts at 5 s; Wan starts at 2 s.

Resolution changes the price much more than duration does. Omni at 360p is the lowest rate in the table; Wan at 1080p is the highest.

Why Veo is the odd one out

Google's Veo page ties its durations to resolution and inputs: 8 s is the only length available with 1080p, 4K or reference images (read 2026-10-04). A 7 s request has no home there.

On Sume, the Video Router catalog is the source of truth for every model's range, so a 7 s request either passes validation or returns a 400 before any money is held. You can pair that with the stored guide to Veo's 4, 6 and 8 second rule when you are choosing between vendors.

What a 7 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.

A 7 s clip at 720p bills $0.88 on Omni and $0.88 on Wan. The Omni 360p draft rate is $0.27 for the same seven seconds.

Check 7 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 7 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 = 7
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"])