Cheapest possible job per Sume video model: shortest clip, lowest tier

The cheapest valid job on each Sume video model is its minimum duration at its lowest resolution: $0.1125 for 3 s on Gemini Omni Flash 1.1 at 360p, $0.125 for 2 s on Wan 3.0 at 480p, $0.3125 for 5 s on MiniMax H3 or H3 Max at 480p, $0.70308 for 4 s on Seedance 2.0 and $1.074708 for 4 s on Seedance 2.5.

Prices on this page are Sume list prices as of 2026-10-08: the provider list price times 1.25, computed from Sume's pricing package and billed per output second, linear in duration inside each model's valid range. The model ids, ranges and input types come from the Video generation docs and the Video Router docs; the per-second numbers can be cross-checked in the public catalog.

Smallest job, and the cost per second of that job

A model with a short minimum can look cheap per second and still be dearer per job than one with a longer minimum. Wan 3.0's 480p rate is $0.0625, the same as H3's, but its 2 s minimum makes the job less than half of the H3 floor.

Using the floor for a probe

A smoke test of your integration (submit, poll, download, check usage.cost) needs one job, and the smallest valid job is the cheapest one. The Omni probe at $0.1125 would run 8,888 times for $1,000, if arithmetic is all you need. Your probe should still use the model and fields you plan to ship, because limits differ per model.

The smallest job is also the best check that a reserve and refund path works on your account, as the cost of a mistake is cents.

Scaling this plan up or down

As a yardstick, the plan above is built on Wan 3.0 at 480p, $0.0625 per second. Each extra 5 seconds adds $0.3125, a further $10 of budget buys 160 more whole seconds, and the largest single job the model accepts (30 s) holds $1.875 at submit. The shortest one (2 s) holds $0.125.

Those three numbers are enough to rescale the plan without a new table. If the plan doubles, double the totals; if a clip is shortened, subtract the seconds multiplied by the rate; and if the tier changes, swap the rate for the one in the tables above.

• Per second: $0.0625
• Per 5 s: $0.3125
• Per 10 s: $0.625
• Per 30 s or the model maximum (30 s): $1.875

Check the model's fields first

After the first job completes, read usage.cost on the poll response and compare it with the figure in this post. They should agree, because both are the provider list price times 1.25. If your number differs, the usual cause is a different tier or duration than you planned, not a price change; the public catalog at https://api.sume.com/v1/catalog shows the current rates and needs no key.

Submitting

Submit through POST /v1/videos with model, prompt, duration in whole seconds and resolution; read GET /v1/videos/models first, because supported_durations and supported_resolutions differ per model. Sume reserves the price at submit and usage.cost on the poll response is the billable amount, so the figures here are what leaves the balance.