Cost of one finished 60-second ad: four Wan clips, voice, music
A finished 60-second ad built from four 15-second Wan 3.0 clips at 720p costs $7.9678 on Sume with voice, a music bed, one render and captions.

One finished 60-second ad costs $7.9678 on Sume at 720p: four 15-second Wan 3.0 clips ($7.50), a 900-character voice-over ($0.0428), one Lyria music bed ($0.125), one Timeline render ($0.10) and one caption job ($0.20). At 480p it is $4.2178 and at 1080p $15.4678.
Wan 3.0 lists at $0.05, $0.10 and $0.20 per output second at 480p, 720p and 1080p (read 2026-10-07). Sume bills list times 1.25 on video-router models.
The bill, line by line
Clips dominate. The finishing lines together are small, and they do not change with resolution.
| Line | Rate | 480p | 720p | 1080p |
|---|---|---|---|---|
| 4 Wan clips of 15 s | per output second | $3.75 | $7.50 | $15.00 |
| Voice-over, 900 characters | $0.0475 per 1,000 | $0.0428 | $0.0428 | $0.0428 |
| Lyria music bed | $0.125 per audio | $0.125 | $0.125 | $0.125 |
| Timeline render, 1 output minute | $0.10 per minute | $0.10 | $0.10 | $0.10 |
| Standalone captions | $0.20 per job | $0.20 | $0.20 | $0.20 |
| Total | $4.2178 | $7.9678 | $15.4678 |
What changes the total
The clip line is 94 percent of the 720p total. Going from 720p to 1080p adds $7.50; changing the voice-over from 900 to 1,800 characters adds $0.0428.
The render bills per output minute rounded up, so a 61-second ad costs two minutes, or $0.20. Keep the cut at or under 60 seconds. The captions job is also stated for videos of at most 60 seconds, so 60 seconds is the useful ceiling for this recipe on both lines.
Wan 3.0 accepts 2 to 30 seconds per job, so two 30-second clips are also an option. The per-second price is the same, and the total clip line does not change; the difference is how many jobs you track and how much you lose when one fails.
- Draft the four clips at 480p for $3.75, approve, then re-render at 720p.
- Count voice-over characters from the final script, since the rate is per 1,000 characters.
- Reuse the music bed across ads, since its price is flat.
Splitting the clip line
The four clips need not share a resolution. Put the hook and the closing shot at 1080p and the two middle shots at 480p: 30 seconds at $0.25 per second plus 30 seconds at $0.0625 per second is $9.375, between the 480p and 1080p rows above. Check the render with the unbilled plan call before you submit it.
Retakes and the budget line
Add a retake allowance to the clip line, not to the finishing lines. If one of the four clips is regenerated once at 720p, add $1.875 and the ad becomes $9.8428. Two retakes bring it to $11.7178.
Fifty such ads at the no-retake price come to $398.3875. Write the retake rate you expect next to that figure before you submit the batch.
Before you submit
Compute the ad from the table, run the unbilled plan call on the render, and send a dry_run for the first clip. Then submit the four clips, wait for them to complete, and only then pay for the render and captions. If a clip fails it is refunded, and you have not yet spent the finishing lines on an ad that cannot be built.
Per finished minute
Because this ad is 60 seconds, the price per finished minute equals the ad total: $7.9678 at 720p. For shorter ads the finishing lines stay the same, so a 15-second ad costs $2.3107 at 720p, of which $0.4357 is finishing. See two 30-second jobs or four 15-second jobs and finishing 100 holiday ads for neighbors.
Sources
Related posts
More in Pricing
- Price per kept AI clip when 1 in 3 is rejected: Wan 720p
A 15-second Wan 3.0 clip at 720p is $1.875 on Sume. If you keep one clip in three, the real price per kept clip is $5.625. The keep rate is a budget input.
- 8-second 720p Seedance price ladder: mini, fast, 2.0 and 2.5 on Sume
One 8-second 720p clip costs about $1.51 on seedance-2-mini and $4.62 on seedance-2.5. The token formula behind each Sume price, with the 1.25 margin shown.
- GPT Image 2.5 at 4:5 lists 12-14% below 1:1 at the same quality
On Sume, GPT Image 2.5 at 4:5 (1024x1280) lists 12 to 14 percent below a 1024x1024 square at every quality tier. The five-tier table, with billed prices.
- GPT Image 2.5 quality ladder: price multiplier from low to max on Sume
At 1024x1024 on Sume, GPT Image 2.5 high costs 8.9 times low, xhigh 15.9 times and max 35.7 times. The ladder with billed prices and batch costs.
Written by Sume