A 150-second Shorts episode from five 30-second AI jobs: cost

A 150-second episode needs five 30-second jobs, which is $18.75 on Wan 3.0 at 720p and $86.67 on Seedance 2.5 at 720p, plus a small join step. No single Sume video request exceeds 30 seconds, so a long Shorts episode is always a chain of clips.

YouTube allows episodes up to 180 seconds in a Shorts series (YouTube Help, read 2026-10-07). The longest single clip Sume accepts is 30 seconds on wan-3.0 and seedance-2.5 (Video generation), so 180 seconds is six jobs and 150 is five.

What five jobs cost

Sume bills the provider list rate multiplied by 1.25 on every video model (Video Router). The figures here come from the price estimators in Sume's repository, run on 2026-10-07, and are rounded to the cent. The live numbers for your workspace are in the catalog response and on the API pricing page.

Six jobs for the full 180 seconds

At the three-minute ceiling the numbers scale by 6/5: $22.50 on Wan 3.0 at 720p, $45.00 at 1080p, and $104.00 on Seedance 2.5 at 720p. A full 180-second season of eight episodes on Wan 720p is $180.00.

Joining the clips

Clips join in one Timeline 1.0 render (Timeline docs). The public rate is $0.10 per started output minute, so a 150-second episode adds $0.30 and barely moves the total.

Plan the cut points before you generate: end each clip on a motion you can hold, and give the next clip a first frame taken from the last one. Use the same ratio and resolution on every segment so the join needs no rescaling.

• Generate the five clips as separate jobs with separate idempotency keys.
• Check each result before the next one is paid for if the story depends on continuity.
• Do not mix 720p and 1080p segments in one episode.

Budget the reserve too

Sume reserves the estimated amount when it accepts a job, captures it on success, and refunds the reservation if the job fails or is cancelled before capture (Usage, Generation admission). So the balance you need on hand at submit time is the price of everything in flight, not only the final total.

A Pro workspace runs 4 jobs at once, Startup 8 and Scale 20 (the plans page lists the ladder), so a big batch drains in waves rather than all at once.