Do Omni references cost extra? Ten images bill like none
On Sume, Gemini Omni Flash 1.1 bills per output second by resolution. A clip with 10 reference images and 3 clips costs the same as one with none.
No. On Sume, a Gemini Omni Flash 1.1 request with ten reference images and three reference videos bills the same as a text-only request of the same length and resolution. The Video Router docs describe the charge as the provider list times 1.25 per output second, as a function of the resolution, and list no per-reference fee.
Same clip, different inputs
The table prices one 8-second clip at 720p (8 x 12.5 = 100 cents) across input shapes the model accepts.
Sume rounds each clip up to the next whole cent, so a clip total is ceil(seconds x rate), not a sum of fractional cents.
| Input shape | Reference fields | Mode | Billed |
|---|---|---|---|
| Prompt only | none | text_to_video | $1.00 |
| Start frame | image_url | image_to_video | $1.00 |
| One reference image | reference_image_urls (1) | reference_to_video | $1.00 |
| Ten reference images | reference_image_urls (10) | reference_to_video | $1.00 |
| Images and clips | 10 images, 3 clips | reference_to_video | $1.00 |
Why this is not true everywhere
Do not generalize. Other Sume video models price differently: the Video generation docs show a per-1,000-video-tokens SKU for Seedance 2. Each row of the catalog has its own pricing block, so read it before you copy a budget from one model to another. For Omni the pricing basis is per second by resolution and nothing else.
That makes Omni a good fit for reference-heavy work like a product shown from several angles. Extra references cost you only the time to prepare them, and the price of a retry is the price of the clip.
What can still raise the bill
Three things change the total. A longer duration, a higher resolution, and a retry that you submit under a new Idempotency-Key. A replay with the same key returns the original job, so retries after a network error do not bill twice.
Read usage.cost on the completed job to confirm. If it ever differs from the table, trust the poll response and the live catalog over this page.
The same length at every tier
For reference, a 8-second Omni Flash clip at each resolution. Every price is the seconds times the billable rate, rounded up to a whole cent, as of 2026-10-08.
| Resolution | Arithmetic | Billed |
|---|---|---|
| 360p | 8 x 3.75 = 30 cents | $0.30 |
| 720p | 8 x 12.5 = 100 cents | $1.00 |
| 1080p | 8 x 18.75 = 150 cents | $1.50 |
| 4K | 8 x 37.5 = 300 cents | $3.00 |
Limits to remember
These apply to every request on this page, from the Video Router and Video generation docs:
- Length is 3 to 10 whole seconds in generation modes; an edit takes no duration.
- Aspect ratio is 16:9 or 9:16; an edit takes no aspect ratio.
- Native synced audio is always on, and
generate_audio: falseis rejected. - There is no
bitrate_mode, no reference audio and noseed. - Billing is the provider list times 1.25 per output second, reserved at submit and shown in
usage.cost.
Sources
Related posts
More in Pricing
- Does Seedance aspect ratio change the price? 21:9 vs 9:16 on Sume
Aspect ratio barely moves Seedance prices on Sume: 9:16, 16:9 and 1:1 cost the same and 21:9 or 4:3 add under a percent. The 720p table and the pixel sizes.
- Does 9:16 cost less than 16:9 for a 30 s Seedance 2.5 clip?
No: 9:16 and 16:9 cost the same on Seedance 2.5 and Wan 3.0 at every resolution. Only 1:1 at 480p is cheaper, by 36 cents on 30 s.
- Editing a minute of video: AA board prices run $9.60 to $40.82
Nine entries on the AA editing board list $9.60 to $40.82 per minute of footage. Sume's one edit model, Gemini Omni Flash 1.1, is $7.50 a minute at 720p.
- Eight 0.5K drafts, then one 4K final: Nano Banana 2.1 costs $0.80
Iterate a hero image eight times at 0.5K for $0.60, then render the keeper at 4K for $0.20: $0.80 total on Nano Banana 2.1 through Sume, versus $1.60.
Written by Sume