Wan 3.0 vs Seedance 2.5 at 30 seconds: the price gap on Sume

A 30-second 720p clip costs $3.75 on Wan 3.0 and about $17.33 on Seedance 2.5 on Sume, a gap of 4.6x. A table across resolutions and what else differs.

5 min readSume
All posts

Thirty seconds at 720p costs $3.75 on Wan 3.0 and $17.33 on Seedance 2.5 on Sume, so Seedance is about 4.6 times the price. Both accept the full 30 seconds and both generate audio when asked.

Price is not the whole story: the two models differ in reference limits and minimum length, covered below.

Price table

30 s, 16:9, Sume price (read 2026-10-07)
ResolutionWan 3.0Seedance 2.5Ratio
480p$1.88$8.064.3x
720p$3.75$17.334.6x
1080p$7.50$42.655.7x

Other differences in the catalog

Sume catalog fields (read 2026-10-07)
Fieldwan-3.0seedance-2.5
Duration2 to 30 s4 to 30 s
Images as referencesup to 10up to 9
Video referencesup to 5up to 3
Audio referencesup to 5up to 3
Billing basisper second by resolutionper 1,000 video tokens

Reading the gap

Wan is the cheaper way to test a 30-second idea. Seedance 2.5 is the pick when you want ByteDance's long-take referencing behavior and are willing to pay for it. Run both on one prompt at 480p first; at 480p and 30 seconds Seedance costs about $6.19 more.

Where the gap comes from

Wan 3.0 is billed per output second from fal's list, with 720p at $0.10. Seedance 2.5 is billed on video tokens at $0.0214 per 1,000, and a 30-second 720p 16:9 clip is 648,000 tokens, which is about $13.87 at list before Sume's 1.25 factor.

The point is structural: token billing makes resolution and duration both expensive, while Wan's flat rate stays low.

What you give up and gain

The cheaper model is not strictly worse. Wan 3.0 accepts more references per job and a shorter minimum clip. Seedance 2.5 is the model ByteDance built for long takes, and its launch post stresses referencing across clips.

Because we have no vendor benchmark on this page to cite, test with your own brief. A paired run at 480p costs a few dollars.

Sources

Related posts

More in Comparisons

All Comparisons posts

Written by Sume