24 seconds: one Wan 3.0 job or three 8-second Omni clips, priced

At 720p, 24 seconds costs $3.00 whether it is one Wan 3.0 job or three 8-second Omni Flash clips. At 1080p the three Omni clips are $4.50 against $6.00 for one Wan job, and at 480p one Wan job is $1.50.

Side by side

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.

Wan 3.0 accepts 2 to 30 seconds in one request, Omni Flash 3 to 10 (Video generation), so 24 seconds is one job on Wan and at least three on Omni.

What differs besides price

One Wan job is a single continuous clip. Three Omni clips are three shots that you join, which suits an edit with scene changes and gives you a retake of one 8-second beat for $1.00 at 720p instead of a whole 24-second retake at $3.00.

Omni always generates native synced audio and accepts image and video references; Wan 3.0 accepts audio, image and video references (Video generation).

• Continuous action in one take: Wan.
• Three discrete beats with 1080p: Omni.
• Need a given audio track to drive the clip: Wan, which takes audio references.

Joining and reserve

Join the three clips with a Timeline render, which is billed per output minute at a small rate next to generation (API pricing). 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.

Set a cap before you start: from MCP, a paid tool takes dry_run=true and max_spend_usd (MCP tools and gates). Send a distinct Idempotency-Key per job so a retry replays the original job instead of charging again.