Why a 3-second Gemini Omni clip bills 12 cents, not 11, on Sume
Sume rounds each video job up to the cent after the 1.25 multiple: a 3-second 360p Omni Flash clip is $0.12, a 3-second 4K clip $1.13. Row-by-row math.

If you multiply by hand, a 3-second Gemini Omni Flash 1.1 clip at 360p looks like $0.09 of provider cost times 1.25, or $0.1125. The Sume invoice line says $0.12. That is not an error: the Video Router docs bill list times 1.25, and Sume's pricing code rounds each job's billable amount up to the next cent.
The four resolutions at the minimum length
Omni Flash 1.1 on Sume runs 3 to 10 seconds. The list rates are the fal prices of $0.03, $0.10, $0.15 and $0.30 per second at 360p, 720p, 1080p and 4K (fal page read 2026-10-05). The table shows the shortest allowed clip and an 8-second clip, which is Sume's default length for sume/auto.
| Resolution | List per second | 3 seconds | 8 seconds |
|---|---|---|---|
| 360p | $0.03 | $0.12 | $0.30 |
| 720p | $0.10 | $0.38 | $1.00 |
| 1080p | $0.15 | $0.57 | $1.50 |
| 4K | $0.30 | $1.13 | $3.00 |
Why it matters for batches
Rounding is per job, not per second. A hundred 3-second 720p clips are a hundred $0.38 lines, which is $38.00, not the $37.50 you get from 100 x 3 x $0.125. The gap is small, but a finance export that reconciles to the cent will show it.
- Budget with the rounded per-clip figure, not per-second x seconds.
- Read
usage.coston each completed job; it is the Sume billable amount. - One 8-second clip at 720p is exactly $1.00 because 8 x $0.125 needs no rounding.
Request
The call is the same one you use for any catalog model. Send model: "gemini-omni-flash-1.1", a duration of 3, and resolution of 360p. The default resolution is 720p.
curl -X POST https://api.sume.com/v1/video-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: omni-3s-360p-001" \
-d '{
"model": "gemini-omni-flash-1.1",
"prompt": "A paper boat crossing a puddle, rain sounds",
"duration": 3,
"resolution": "360p",
"aspect_ratio": "16:9",
"mode": "async"
}'
Sources
Related posts
More in Pricing
- How Sume pricing works: plans, one wallet, published model rates
Sume plans set access and concurrency. Usage draws from one prepaid wallet at each model's published USD rate, for generation, the Agent, Formats, and the API.
- AI avatar video API pricing: cost per second and per minute
Sume bills AI avatar video per second by quality tier, with separate rates when you send a product image. Per-minute costs for standard, plus, and max.
- AI video generation cost per video: what one Sume API run cost
To see what one Sume run or video job cost, call GET /v1/usage with run_id or job_id and read debited_usd: the wallet deduction, agent turns included.
- Do failed AI video generations cost credits? Reserve, capture, refund
No. Sume reserves a job's estimated USD cost at submit, captures usage only on completion, and releases or refunds the hold if the job fails or is canceled.
Written by Sume