Swap the gift in a Christmas clip with Gemini Omni Flash edit
Gemini Omni Flash 1.1 on Sume edits a 3 to 10 second clip by prompt, so one approved Christmas ad can become three gift variants without a reshoot.

Gemini Omni Flash 1.1 on Sume accepts a video_url and a prompt and rewrites the clip, for example replacing the bottle with an apple. That lets one approved Christmas ad become three gift variants. The input is 3 to 10 seconds, so trim a longer clip first.
What the edit mode does
You send an existing clip and describe the change. The model keeps the framing and motion and swaps the object. It does not take reference photos in edit mode, so describe the new gift in words.
| Field | Value |
|---|---|
| Clip length | 3 to 10 seconds |
| Edit input | video_url plus a prompt |
| Example prompt | replace the bottle with an apple |
| Billing | provider list times 1.25 |
Three variants, one source
Write three prompts that change only the object: candle, mug, scarf. Run each as a separate job and review the three clips side by side before captions.
- Keep the prompt short and literal
- One change per prompt
- Check hands and edges frame by frame
When to trim first
If your source clip runs over 10 seconds, cut the best 8 with Sume's video trim at $0.02 per job, then edit the trimmed copy.
Limits worth knowing
A product with a legible logo or label may not survive an edit cleanly. For hero shots with exact packaging, generate from a first frame instead, and keep edit mode for non-label objects.
Sources
Related posts
More in Models
- Does Lyria 3.5 audio carry a SynthID watermark?
Yes: Google's Gemini API music docs say Lyria audio carries a SynthID watermark. The Sume docs I read do not describe watermarking either way.
- Eleven v4 stacked tags: direct emotion without them
ElevenLabs v4 adds stackable expression tags and 10-second cloning. Sume's docs list no clone route; here is how to direct a voice via tts_create.
- eleven_v4 vs eleven_v4_turbo: model IDs, endpoints, which to pick
ElevenLabs lists eleven_v4 for expressive speech with cloning in 90+ languages and eleven_v4_turbo at about 100 ms median latency. Which fits a video pipeline.
- ElevenLabs languages: Flash v2.5 has 32, Multilingual v2 29, v4 90+
ElevenLabs lists 32 languages for Flash v2.5, 29 for Multilingual v2 and 90+ for v4 and v4 Turbo. Check your markets against the model, then log it per job.
Written by Sume