Veo 3.1 Fast vs Omni Flash: both $0.10 a second at 720p
Google lists Veo 3.1 Fast and Gemini Omni Flash near $0.10 a second at 720p. Sume lists Omni only, at $0.125. Here is the 8-second math and what differs.

On Google's own pricing page, Veo 3.1 Fast is $0.10 a second at 720p and Gemini Omni Flash (gemini-omni-1.1-flash) is billed at about $0.10 a second of 720p video, so an 8-second clip is about $0.80 either way. Sume lists gemini-omni-flash-1.1 at $0.125 a second at 720p ($1.00 for 8 seconds), and Sume does not list Veo 3.1 at all.
What Google's page says
Google lists Veo 3.1 in three tiers. The page also gives Omni Flash an output rate for video and states about $0.10 a second at 720p. The Omni figure is a derived approximation on the page, so treat it as a planning number.
| Option | 720p per second | 1080p per second | 8 seconds at 720p |
|---|---|---|---|
| Veo 3.1 Lite (Google) | $0.05 | $0.08 | $0.40 |
| Veo 3.1 Fast (Google) | $0.10 | $0.12 | $0.80 |
| Veo 3.1 Standard (Google) | $0.40 | $0.40 | $3.20 |
| Gemini Omni Flash (Google) | about $0.10 | not stated | about $0.80 |
| gemini-omni-flash-1.1 (Sume) | $0.125 | $0.1875 | $1.00 |
The Sume row
Sume bills the provider list times 1.25, which gives $0.125 at 720p (0.10 x 1.25) and $0.1875 at 1080p. The model accepts 3 to 10 seconds, 16:9 or 9:16, and native synced audio is always on: the API rejects generate_audio false. It also takes reference images (up to 10) and short reference videos, and it can edit an existing video. None of those capabilities are claimed for Veo here, because this post only reads Google's price page for Veo.
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-8s-720p" \
-d '{
"model": "gemini-omni-flash-1.1",
"prompt": "A vertical product clip on a desk, natural light",
"resolution": "720p",
"duration": 8,
"aspect_ratio": "9:16",
"mode": "async"
}'How to choose
If you need Veo 3.1 by name, Sume is not the route today. If the requirement is a Google-family video model with built-in audio at about ten cents a second, the Omni row is the one Sume lists, and the premium over Google's own number is the $0.025 a second margin: $0.20 on an 8-second clip.
If audio is a hard no, Omni is out because it is always on. Wan 3.0 at 720p is also $0.125 a second on Sume and takes a 2 to 30 second range.
A 100-clip example
For 100 eight-second clips at 720p, Google's page implies about $80 on Veo 3.1 Fast and about $80 on Omni Flash. On Sume the Omni row gives 100 x 8 x $0.125 = $100.00. The $20 difference is the 25 percent margin. If you hold a Google billing account already, the direct route is cheaper for the same Omni model, and the Sume route buys you the shared job envelope, the shared balance and the other rows in the catalog.
For 1080p, Veo 3.1 Fast is $0.12 a second and Veo 3.1 Lite is $0.08, so 100 eight-second clips are $96 and $64. Sume's Omni 1080p row is $0.1875 a second, which is $150.00 for the same batch. Google's page gives no 1080p figure for Omni Flash in the part read here, so that comparison is left open.
Sources
Related posts
More in Comparisons
- Veo 3.1 Lite: 10 seconds at 720p costs $0.50; Wan 3.0 on Sume $1.25
Veo 3.1 Lite lists $0.05 per second at 720p on Google. Sume does not list Veo; its Wan 3.0 row is $0.125 per second at 720p. The 10-second arithmetic.
- Veo 3.1 Lite 1080p is $0.08/s at Google; H3 Max is $0.20 on Sume
Google lists Veo 3.1 Lite at $0.08 per second at 1080p. Sume lists no Veo; its 1080p rows are H3 Max $0.20, Wan 3.0 $0.25, Omni $0.1875, Seedance 2.0 $0.8505.
- Veo 3.1 takes 3 reference images; Omni on Sume takes 10 plus 3 clips
How many reference images does Veo 3.1 accept, and how many does Gemini Omni Flash take on Sume? Limits, tag syntax and a request example.
- Max AI video length: Runway Gen-4.5 10 s vs Sume's 30-second rows
Runway's Gen-4.5 runs 2 to 10 seconds and Omni 3 to 10. Sume's Wan 3.0 and Seedance 2.5 reach 30. Longest single clip by model, with the sources.
Written by Sume