Flow offers Omni at 360p and 720p; the API lists 1080p and 4K
Flow's help page shows Gemini Omni Flash 1.1 at 360p and 720p; the Gemini API and Sume list 1080p and 4K too. Which one to use depends on the output you ship.

Google Flow's models page lists Gemini Omni Flash 1.1 with 360p and 720p options (read 2026-10-02), while the Gemini API page lists 360p, 720p, 1080p and 4K, with 4K upscaled and 720p the default. Sume's gemini-omni-flash-1.1 lists the same four tiers. Pick by where the clip ends up: a draft in Flow, a delivery file through the API.
Three Google pages, one model
Flow's help table shows Omni Flash 1.1 at 360p and 720p (Flow Help, read 2026-10-02). The Gemini API page lists 360p, 720p, 1080p and 4K (upscaled), 16:9 default and 9:16 (Gemini API docs, read 2026-10-02). Google's developer blog says 360p is up to 60 percent faster and one third the cost of 720p (Google blog, read 2026-10-02).
| Page | Resolutions listed | Notes |
|---|---|---|
| Flow models table | 360p, 720p | Durations 4, 6, 8, 10s |
| Gemini API Omni docs | 360p, 720p, 1080p, 4K | 720p default; 4K upscaled |
| Google developer blog | 360p to 4K | 360p: up to 60% faster, one third the cost of 720p |
On Sume
gemini-omni-flash-1.1 runs 3 to 10 seconds at 360p, 720p, 1080p and 4K, in 16:9 or 9:16 with native audio, through resolution on the video routes (Video generation). Billing is provider list times 1.25 per output second by resolution, and the rate by tier is on the models endpoint.
Edits through video_url default to 720p and take no aspect_ratio or duration.
A draft then final workflow
Run the prompt at 360p first to check motion and audio, then repeat the exact prompt at the delivery tier. Because Sume has no seed field, the second run is a new draw, not the same clip bigger; use the draft to check the idea, not to lock the frames.
If the real need is a still frame for a thumbnail, take it from the final render, not the draft.
Which surface for which job
Flow is where Google surfaces what each plan includes in the app. For output you place into an ad, a storefront or a timeline, the 1080p or 4K tiers come from an API, and you should confirm the tier on that surface before promising it to a client.
Sources
Related posts
More in Models
- Flow Veo 3.1 Quality has no ingredients: references on Sume Omni
In Flow, Veo 3.1 Quality takes text and frames only, while Fast and Lite take ingredients. Sume's reference-to-video runs on Omni with up to 10 images.
- FLUX.2 takes 8 references by API, 10 in Playground: Sume's range
Black Forest Labs says FLUX.2 pro, max and flex take up to 8 references by API (10 in its Playground), klein 4. Sume's input_references range is read per model.
- flux-2-pro-preview vs flux-2-pro: which id does Sume send?
BFL has flux-2-pro-preview (latest) and flux-2-pro (fixed snapshot). Sume lists black-forest-labs/flux.2-pro and does not let you pick the BFL endpoint.
- gemini-omni-1.1-flash vs gemini-omni-flash-1.1: which id goes where
Google writes the model id gemini-omni-1.1-flash; Sume's catalog id is gemini-omni-flash-1.1. A mapping table, the old preview id, and how to avoid a typo.
Written by Sume