Can Gemini Omni edit change a video's aspect ratio? No
On Sume, Omni's edit mode takes a video_url and a prompt, defaults to 720p and accepts no aspect_ratio or duration. A 16:9 source stays 16:9, so plan the crop.

No. On Sume, Gemini Omni Flash 1.1's edit mode keeps the framing of the source video, and the edit request takes no aspect_ratio or duration. A horizontal clip comes back horizontal.
If the end goal is a 9:16 ad, shoot or generate it vertical first, or crop with video filter before or after the edit.
What an edit request takes
Edit is the video_to_video operation on the Video Router. The source goes in video_url, and the prompt carries the instruction. resolution is optional and defaults to 720p. video_url is the edit source, not a reference, and the video generation docs say you cannot send it with the reference fields.
| Field | Edit mode | Note |
|---|---|---|
| video_url | required | the source being edited |
| prompt | required | gives the instructions for the edit |
| resolution | optional | default 720p |
| aspect_ratio | not accepted | framing follows the source |
| duration | not accepted | length follows the source |
Three ways to get a vertical result
Each route changes cost and look. Choose by what the footage can spare.
- Generate vertical from the start. Omni makes 9:16 natively on a text or image prompt. No edit step, no crop.
- Crop first, then edit. Use video filter to cut a 9:16 strip from a 16:9 source (a 0.3164 crop width of the frame), then send the cropped file to the edit. The edit sees the final framing.
- Edit first, crop last. Cheaper to reason about, but the edit spends pixels on parts of the frame the crop throws away.
Edit or regenerate
Edit when you like the composition and need one change: steam on a cup, a different colour, decorations in a shop. Regenerate when you want a new composition or a new aspect ratio, because an edit starts from the source's framing and stays near it.
For a change of ratio the cheapest route is usually a crop, which is a flat $0.02 per job on video filter, not a new generation.
Check before you pay
The edit SKU and its price are listed in pricing_skus on GET /v1/videos/models. I have not quoted an edit price here, because the card shows it per model and resolution and may change. Run one short clip first. A trim at $0.02 gets a test cut out of a long source.
Sume does not guarantee that an edit leaves the rest of the frame untouched. Compare the result against the source, especially product labels and faces.
Sources
Related posts
More in Media tools
- Caption a talking-head video: style, placement and phrasing
Caption a talking-head clip with the Sume API: pick a style, move the line off the face, set words per card. One request, $0.20 for up to 60 seconds.
- Captions for a video with loud background music: script text or cues
Music can bury speech and trip speech-to-text. Sume documents three ways to supply your own wording: script_text, a words array or cues. Which to pick.
- ChatGPT Sora clip as reference video: Omni takes 3 s each, trim
Sora 2 left the API but a ChatGPT clip can still seed a new render. Gemini Omni Flash 1.1 on Sume takes up to 3 reference videos, each 3 s at most.
- Content ID on a Short over one minute blocks it globally: keep a cut
YouTube says a video over one minute with an active Content ID claim is blocked globally. Keep a 59-second video-trim cut of every long Short as a fallback.
Written by Sume