20-second highlight reel: Seedance 2.5 at $5.38 vs Wan 3.0 at $2.50
A 20-second highlight clip costs $5.38 on seedance-2.5 at 480p and $2.50 on wan-3.0 at 720p on Sume. Use video references to keep the look.

A 20-second motion-graphic highlight clip costs $5.38 on seedance-2.5 at 480p and $2.50 on wan-3.0 at 720p on Sume. Both reach 20 seconds in a single job, so there is no stitching on either side.
Neither model reads a webinar recording and cuts highlights from it. They generate new footage from a prompt and references. Use them for the animated intro, section cards and b-roll that surround your real clips.
Price for 20 seconds
Wan 3.0 lists at $0.05, $0.10 and $0.20 per second and Sume bills list times 1.25. Seedance 2.5 is billed per video token times 1.25, which is about $0.269 per second at 480p (a 30-second clip is $8.06) (video router docs).
| Model | Resolution | 20-second clip |
|---|---|---|
| wan-3.0 | 480p | $1.25 |
| wan-3.0 | 720p | $2.50 |
| wan-3.0 | 1080p | $5.00 |
| seedance-2.5 | 480p | about $5.38 |
Where video references help
Both ids accept image, video and audio references (video generation docs). Send a short clip of your brand's motion style as a video reference and a frame of your slide template as an image reference. ByteDance says Seedance 2.5 understands reference videos for their framing and cinematic language, not only their motion (read 2026-10-07, Seed). That is useful here, because the goal is a clip that matches your existing look.
How to assemble
Generate the card or b-roll at 20 seconds, trim to what you need, and cut it between the real speaker segments in your editor. Because you pay per generated second, generate only as long as the cut needs: a 6-second Wan clip at 720p is $0.75, not $2.50.
Decision
Start on Wan 3.0 at 480p ($1.25). If the style does not match, move the same references to Seedance 2.5 at 480p ($5.38) before paying for a higher tier.
Practical notes
A useful split is to use generated clips for transitions and keep the substance real. A 4-second Wan 3.0 sting at 720p is $0.50; use a handful of those between real speaker clips and the generated share of the budget stays small.
Keep a log of which references produced a match. Over a few webinars you will build a small library of reference clips and images that reproduce your look on demand.
Sources
Related posts
More in Comparisons
- Can I use Suno Speech beta for an ad voiceover? What to check first
Suno Speech beta makes one track with voice and music. Before using it for ads, check price, languages and edits, then see how Sume splits voice and music.
- Which TikTok ad placements take a 10-minute video? Only auction
TikTok's auction non-Spark page allows up to 10 minutes; reservation in-feed and TopView stop at 60 seconds. Pick the placement first, then cut with video-trim.
- Zoom in on a small product in frame: AI recompose or crop and upscale
Product too small in the photo? A Pillow crop plus Sume Image Upscale ($0.20) keeps real pixels; an Ideogram 4.5 recompose edit costs $0.075 and redraws them.
- Sume vs Argil: AI avatar video and video agents compared
Argil makes AI-avatar and story videos with a chat agent, Director; Sume is a video agent with a multi-model API. Avatars, API, pricing, and limits compared.
Written by Sume