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.

Google's Veo 3.1 page lists up to three reference images as style and content guides (read 2026-10-09). On Sume, gemini-omni-flash-1.1 accepts up to 10 reference images and up to 3 reference clips of 3 seconds each, and it takes no reference audio.
What do the two limits look like side by side?
Google's Omni page lists up to 3 video references of 3 seconds each and says audio references are unsupported. Sume's Video Router docs set the same video limits and the 10-image limit.
| Input | Veo 3.1 (Google) | Omni on Sume |
|---|---|---|
| Reference images | Up to 3 | Up to 10 |
| Reference video clips | Not listed on the Veo page | Up to 3, each 3 seconds or less |
| Reference audio | Not listed | Not accepted |
How do you point at a reference in the prompt?
On Sume you name each reference by index, from zero, in list order: <IMAGE_REF_0>, <IMAGE_REF_1>, and <VIDEO_REF_0>. In a /v1/videos call, the images go in input_references; on the Video Router wire they go in reference_image_urls and reference_video_urls.
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: refs-001" \
-d '{"model":"gemini-omni-flash-1.1","prompt":"<IMAGE_REF_0> walks past <IMAGE_REF_1>, handheld","reference_image_urls":["https://example.com/a.png","https://example.com/b.png"],"duration":6,"resolution":"720p","mode":"async"}'Does a reference change the price?
Sume bills Omni per output second by resolution, so the 6-second 720p call above is 6 x $0.125 = $0.75, however many images you send. The cap is the thing to watch: an eleventh image is an error, not a charge.
What if you only need Veo?
Sume does not list a Veo id (catalog read 2026-10-09). Google marks the Veo 3.1 preview ids for shutdown on October 22, 2026, with Omni as the named replacement.
Sources
Related posts
More in Comparisons
- 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.
- Vidu Q4, Wan 3.0 and Seedance 2.5: reference inputs compared
Images, videos and audio you can send as references: Vidu Q4 Preview and Wan 3.0 from their pages, wan-3.0 and seedance-2.5 on Sume from the docs.
- YouTube AI label: three auto-detection signals and Sume outputs
YouTube may label video automatically for its GenAI tools, C2PA metadata or internal detection. Sume docs do not mention C2PA, so here is what to check.
- Wan 3.0 vs Seedance 2.5: the keep rate where Wan is cheaper
Wan 3.0 at 720p costs $0.63 per 5 s and Seedance 2.5 costs $2.89 on Sume. Wan wins per kept clip unless its keep rate falls below about 21.8% of Seedance's.
Written by Sume