Car lot video: 60 vehicles, three 8-second clips each, on Wan 3.0
Photos to video for 60 vehicles with Wan 3.0 at 8 seconds a clip: $97.20 at 480p, $187.20 at 720p, $367.20 at 1080p with the render and a voice line.

Turning the photos of 60 vehicles into three 8-second clips each costs $187.20 on Wan 3.0 at 720p on Sume: 180 clips at $1.00, a $0.10 timeline render per vehicle for the 24-second cut, and a $0.02 voice line per vehicle (a 300-character TTS job). At 480p the same lot costs $97.20 and at 1080p $367.20. Prices are Wan 3.0's list rate times 1.25, rounded up to the cent for each job (read 2026-10-06).
Wan 3.0 accepts 2 to 30 seconds per clip, three resolutions, and image-to-video as well as text-to-video, so a dealership can start from the listing photos it already has. The table prices the lot at each resolution and then prices the monthly refresh, when only 15 of the 60 vehicles are new.
| Resolution | One 8-second clip | Three clips | Per vehicle with render and voice | 60 vehicles | 15 new vehicles a month |
|---|---|---|---|---|---|
| 480p | $0.50 | $1.50 | $1.62 | $97.20 | $24.30 |
| 720p | $1.00 | $3.00 | $3.12 | $187.20 | $46.80 |
| 1080p | $2.00 | $6.00 | $6.12 | $367.20 | $91.80 |
What each part of the bill is
The clips are 96 percent of the 720p total: $180.00 of $187.20. The render and the voice are rounding errors by comparison, so the decision that matters is the resolution. Doubling from 480p to 720p doubles the clip price, and 1080p doubles it again, so a lot that ships to a web listing at 720p should not be rendered at 1080p unless a channel demands it.
The render charge is per started output minute, and a 24-second cut is one minute. Three 8-second clips are a safe length: you would need a 61-second cut before the render doubled.
- Use one source photo per clip, and name the camera move in the prompt so three clips of one car do not repeat each other.
- Render at 480p first for one vehicle, review it, and then commit the lot at the chosen resolution.
- Read
GET /v1/balancebefore the batch. A lot of 60 vehicles at 720p needs $187.20 available, and a balance below one clip's price returns402 insufficient_credits. - Failed clips are refunded, so a bad photo costs you a retry, not a second charge.
Price the lot in code
The function below prices any lot size, clip count and resolution from the billed 8-second prices in the table, including the render minute and the voice line.
CLIP_8S = {"480p": 0.50, "720p": 1.00, "1080p": 2.00}
RENDER_PER_MIN = 0.10
VOICE_300_CHARS = 0.02
def lot_cost(vehicles: int, clips_each: int, res: str) -> float:
seconds = clips_each * 8
minutes = -(-seconds // 60)
per_vehicle = (
clips_each * CLIP_8S[res] + minutes * RENDER_PER_MIN + VOICE_300_CHARS
)
return round(vehicles * per_vehicle, 2)
for res in CLIP_8S:
print(res, lot_cost(60, 3, res), lot_cost(15, 3, res))Spend the resolution where it sells
Not every vehicle needs the same tier. Render the 10 featured vehicles at 1080p, $6.12 each or $61.20, and the other 50 at 480p, $1.62 each or $81.00. The lot costs $142.20, which is $45.00 less than rendering all 60 at 720p, and the cars the dealer cares about most get the sharpest clips. You can also render everything at 480p first, then promote only the vehicles that get enquiries.
The mixed plan also gives you a natural retake budget. If five of the 480p clips come out wrong, five retries cost $2.50 in clips, so you can keep a $10 reserve on top of the plan and still land below the all-720p number. Put the plan into a CSV with one row per vehicle and a tier column, and the batch script reads the price from the table instead of from memory.
Keeping the lot fresh
A lot changes weekly, so the number to watch is the refresh. At 720p, 15 new vehicles a month is $46.80, and a sold vehicle is simply not re-rendered. Keep the clip URLs with the vehicle record, so a re-listed car reuses its clips and costs nothing.
If you want one hero clip per vehicle, one 8-second clip at 720p is $1.00 and the lot is $60 plus the voice lines. The live per-model limits and prices are in the Video Router catalog, and the render rate is in the timeline docs.
Sources
Related posts
More in Pricing
- Cheapest AI video clip on each Sume model: the shortest legal length
The smallest legal clip on nine Sume video models: Wan 3.0 at 2 s is $0.13, Omni 3 s $0.12, Seedance 2 Mini 4 s $0.36, H3 5 s $0.32, Seedance 2.5 4 s $1.08.
- Cheapest vs priciest AI video on Sume: a per-second spread of 114x
Sume's video rates run from $0.0125 per second (Grok Imagine Video 1.5) to $1.42155 (Seedance 2.5 at 1080p): the ratios, and what each size buys.
- 40 speaker intro videos: avatar creation plus 15 seconds each on Sume
A 15-second avatar intro for each of 40 speakers costs $148.40 on Standard, $185.20 on Plus and $368.00 on Max, with the $0.95 avatar creation per speaker.
- Cost of one finished minute of AI video by model, with 1, 2 or 3 takes
What 60 seconds of finished AI video costs on Sume at 720p or 768p across Seedance 2.5, Wan 3.0, Omni Flash and MiniMax H3, when you keep every take or 1 in 3.
Written by Sume