Introduction
Gemini Omni 1.2 Flash video-edit begins with footage you already have. Supply video_url, a required prompt and an output resolution. In our tests, changing a gray studio into a moonlit lakeside stage preserved the dancer and movements while replacing the setting.
The edit follows source timing, with footage beyond 10 seconds cut away. In our tests, a 12-second source produced a 10.03-second result. There is no duration or aspect-ratio field for editing. Billing rounds source seconds up, with a minimum of three and a maximum of ten; the selected resolution sets an edit rate higher than generation.
Key features & highlights
- Apply a written visual change to an existing video rather than generating from a still.
- Follow source duration, with a 10-second cut for longer footage.
- Select 360p or 720p direct output, or higher-rate 1080p/4K upscaling by Google.
- Receive an MP4 result at 24 fps with an AAC audio track.
- Keep the request focused using Google’s recommendation to preserve everything outside the change.
Request parameters
| Parameter | Required | Description |
|---|---|---|
prompt | Yes | Required short instruction describing the visual change. Google recommends simple wording ending with “Keep everything else the same.” at least 1 characters · at most 20000 characters |
video_url | Yes | Required source clip URL. Output follows source duration, with sources longer than 10 seconds cut at 10 seconds. at least 1 characters |
resolution | No | Use 360p, 720p (default), 1080p or 4K. Google lists 360p/720p as direct output; 1080p and 4K are upscaled by Google and cost more. |
Pricing
Editing charges source seconds rounded up, with a 3-second minimum and a 10-second maximum. Resolution sets the rate; the edit rate is higher than the generation rate. Sources exceeding 10 seconds are cut at 10 seconds. Failed or refused edits incur no charge.
| Specification | Price |
|---|---|
| $0.0576 / second |
| $0.1248 / second |
| $0.1752 / second |
| $0.3264 / second |
Use cases
- Replace a gray rehearsal studio with a moonlit stage beside a lake.
- Review a misty setting variation while asking to preserve the recorded dance movement.
- Try a different visual environment for an existing short performance before choosing a version.
Prompt guide & pro tips
- Begin with a single change, such as “turn the studio into a lakeside stage,” following Google’s guidance on simple edit prompts.
- Describe a specific visual quality, such as “thin mist on the water,” so the new setting has a clear direction.
- Use Google’s suggested ending “Keep everything else the same.” to limit the requested revision.
- Name an element to preserve, such as “retain the original movement,” when reviewing whether the performance carries through.
- Prepare the intended segment before uploading; for a direction such as “change this scene to moonlight,” content after the first 10 seconds will be cut.
API quick start
# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
-H "Authorization: Bearer $SPICY_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
-d '{
"model": "google/gemini-omni-1.2-flash/video-edit",
"input": {
"prompt": "A slow aerial shot drifting over a misty pine forest at sunrise",
"resolution": "720p",
"video_url": "https://example.com/reference.mp4"
}
}'
# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
-H "Authorization: Bearer $SPICY_API_KEY"
See the API reference for authentication, the complete schema, polling, and error handling.
