Most expensive single job per Sume video model: longest, top tier

The largest single job on each model is its maximum duration at its top resolution: $42.6465 on Seedance 2.5 (30 s, 1080p), $12.7575 on Seedance 2.0 (15 s), $7.50 on Wan 3.0 (30 s), $3.75 on Gemini Omni Flash 1.1 (10 s, 4K), $3.00 on H3 Max (15 s, 1080p) and $1.125 on H3 (15 s, 768p).

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.

Why it matters

Sume reserves the price at submit, so the ceiling of a job is the balance you need to hold to submit it. A workspace with $10 free can submit any Wan 3.0 job, but not a 30 s Seedance 2.5 1080p job.

Reading the spread

The ratio between the top and bottom of the table is 37.9 ($42.6465 against $1.125). A budget guard that reads from supported_durations and the catalog price can refuse a job over a limit you set, before it reaches the API.

Three lines of arithmetic make that guard: look up the per-second price, multiply by the requested duration, compare with your ceiling. Keep the price table in config, because the catalog changes.

Scaling this plan up or down

As a yardstick, the plan above is built on Seedance 2.5 at 1080p, $1.42155 per second. Each extra 5 seconds adds $7.10775, a further $10 of budget buys 7 more whole seconds, and the largest single job the model accepts (30 s) holds $42.6465 at submit. The shortest one (4 s) holds $5.6862.

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: $1.42155
• Per 5 s: $7.10775
• Per 10 s: $14.2155
• Per 30 s or the model maximum (30 s): $42.6465

Check the model's fields first

Put the price table in your own config and compute the total before you call the API. The price is linear in seconds inside a model's range, so a function of rate and duration is enough, and it lets you refuse a batch that goes over a ceiling. Re-read the catalog when you plan a large run, because list prices are what Sume bills from, and they can change.

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.