Does native audio cost extra on Wan 3.0 and Gemini Omni Flash on Sume?
On Sume, Wan 3.0 costs the same with the audio flag on or off, and Gemini Omni Flash 1.1 always includes audio. Billed prices for both, from Sume's code.
No. On Sume, the generate_audio flag does not change a Wan 3.0 quote, and Gemini Omni Flash 1.1 has no audio switch because every clip includes audio. You pay by seconds and resolution for both.
What the code does
Sume's Wan 3.0 estimate reads the audio flag but prices only resolution and seconds. The Omni Flash estimate has the same shape: per-second tiers at 360p, 720p, 1080p and 4K. The Video Router docs say Omni always produces audio, rejects generate_audio: false, and runs 3 to 10 seconds per clip.
This is a property of Sume's price tables, not a general rule. Other models in the catalog price audio differently, so read each model's quote before assuming the same.
Same price, flag on or off
| Model | Seconds | Audio off | Audio on |
|---|---|---|---|
| wan-3.0 | 5 | $0.63 | $0.63 |
| wan-3.0 | 10 | $1.25 | $1.25 |
| wan-3.0 | 30 | $3.75 | $3.75 |
| gemini-omni-flash-1.1 | 3 | n/a, always on | $0.38 |
| gemini-omni-flash-1.1 | 8 | n/a, always on | $1.00 |
| gemini-omni-flash-1.1 | 10 | n/a, always on | $1.25 |
Practical effect
There is nothing to save by turning audio off on these two, so leave it on if you might use the soundtrack and drop it in the edit if you do not. If a silent clip is a hard requirement, strip the track afterward rather than hoping for a discount.
The two models also differ in length: Wan 3.0 reaches 30 seconds, Omni 10. The Video Router docs have the per-model limits, and Wan's per-resolution prices are in their own table.
- Wan 3.0: audio flag has no price effect.
- Omni Flash 1.1: audio always on.
- Check other models separately.
How to verify it yourself
Call the estimate twice for the same Wan 3.0 request, once with generate_audio true and once false. The billed amounts match in the tables above, for 720p.For Omni there is nothing to toggle, so the check is to compare tiers instead.
One caution is that a price rule can change in a later release. This page reflects the pricing code as read on 2026-10-06. If you rely on a flat audio price for a large batch, repeat the two-estimate check first; it takes a minute and costs nothing, because an estimate does not create a job. The API pricing page is the public reference for the current list.
How these numbers were produced
The figures are Sume's billed-amount estimates for wan-3.0 and gemini-omni-flash-1.1 on 2026-10-06, for 720p jobs with the audio flag on and off. The tables show whole cents.
These are estimates of the billed amount for the stated settings. Anything not in the table, such as a different frame shape, a reference input or a changed quality tier, can change the number, so run the estimate for your own request before a large batch. Where this page and the public API pricing list disagree, the pricing page and the live quote win.
Sources
Related posts
More in Pricing
- Estimate a folder of clips transcription cost with ffprobe
Sum each clip's seconds with ffprobe, flag anything over 600 s, and price it at $0.01 per audio minute before you submit a single Sume STT job.
- 5-second Wan 3.0 clip at 480p: 32 cents, not 31, and why
Sume rounds each video job's billable amount up to the cent. A 5 s Wan 3.0 clip at 480p is $0.3125 at list x 1.25, so you are billed $0.32.
- Gemini Omni 1.1 Flash 1080p and 4K price: upscales on Sume
Google says Omni 1.1 Flash renders 720p and offers 1080p and 4K as upscales. Here is what Google lists and what the Sume 10-second clip costs per tier.
- Omni Flash on Sume: cost of 10, 30 and 60 seconds at every resolution
On Sume, Gemini Omni Flash costs $0.375 to $3.75 for 10 seconds, $1.125 to $11.25 for 30, and $2.25 to $22.50 for 60. Full 360p to 4K table.
Written by Sume