13-second AI video API: MiniMax H3 at 768p and other Sume ids

Thirteen seconds fits MiniMax H3 and H3 Max (5-15 s), Wan 3.0, Kling 3 and Seedance. Billed price per resolution for a 13 s clip on Sume.

5 min readSume
All posts

A 13 second clip is a good fit for MiniMax H3 on Sume. Both H3 ids take 5-15 s, and H3 ships native stereo audio on every clip. MiniMax's own announcement lists a 15 s maximum, 2K by default and 768p as the other option (read 2026-10-04).

Which models accept 13 seconds

The Sume catalog lists native 480p and 768p for H3, with 2K and 4K offered as upscales of 768p that are priced only when requested. Sume's default for H3 is 768p, not 2K.

Models whose supported_durations include 13 s on Sume, with billed price for 13 s (provider list x 1.25, rounded up to the cent). Sume docs and code, read 2026-10-04.
Model idDuration rangeBilled for 13 s
minimax-h35-15 s480p $0.82, 768p $0.98
minimax-h3-max5-15 s480p $0.82, 768p $1.30, 1080p $2.60
wan-3.02-30 s480p $0.82, 720p $1.63, 1080p $3.25
kling-34-15 saudio off $1.82, audio on $2.73
seedance-2.54-30 stoken priced, read pricing_skus
seedance-2, seedance-2-fast, seedance-2-mini4-15 stoken priced, read pricing_skus
grok-imagine-video-1.54-15 sread pricing_skus
higgsfield-genjutsu4-30 s (source length)read pricing_skus
h3-max-recast5-30 s (source length)768p $4.88, 1080p $7.32

Thirteen seconds with native audio

Because audio is always on for H3, there is no audio switch to price; Kling 3, by contrast, has separate rates for audio off and on. For a 13 s clip that needs a voice and room tone in one shot, H3 is the simplest path.

H3 accepts reference images, videos and audio, with its own limits. Those are covered in the stored H3 reference limits post.

What a 13 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.

Thirteen seconds of H3 at 768p bills $0.98 and at 480p bills $0.82. H3 Max at 1080p bills $2.60.

Check 13 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 13 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 = 13
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"])

Sources

Related posts

More in Models

All Models posts

Written by Sume