Gemini Omni Flash

Gemini Omni Flash is Google’s Gemini Omni 1.1 Flash video model. Use gemini-omni-flash for text-to-video, first-frame image-to-video, or reference-to-video, at 720p, 1080p, or 4K, for 4, 6, 8, or 10 seconds. It is billed per request, not per second.

`type`	`image_urls`
`text-to-video`	Omit
`image-to-video`	Exactly 1 image, used as the first frame
`reference-to-video`	1 or 3 reference images (2 is rejected)

type is optional. When omitted, it is inferred from the image count: none → text-to-video, 1 → image-to-video, 3 → reference-to-video. Send type: "reference-to-video" explicitly to use a single reference image.

Request parameters

Parameter	Required	Accepted values
`prompt`	Yes	Non-blank text. Describe the scene, subject, action, camera, and sound
`type`	No	`text-to-video`, `image-to-video`, or `reference-to-video`
`image_urls`	For image and reference modes	Public HTTP(S) URLs; count depends on `type`. Base64 is not supported
`aspect_ratio`	No	`16:9` (default) or `9:16`
`resolution`	No	`720p` (default), `1080p`, `4k` (lowercase)
`duration`	No	`4`, `6` (default), `8`, or `10` seconds

Each request produces one video. Reference videos are not offered.

{
  "prompt": "A creative scene combining these elements with smooth camera motion",
  "type": "reference-to-video",
  "image_urls": [
    "https://example.com/scene.jpg",
    "https://example.com/character.jpg",
    "https://example.com/product.jpg"
  ],
  "aspect_ratio": "9:16",
  "resolution": "1080p",
  "duration": 10
}

Submit to POST /api/v1/task/submit/gemini-omni-flash. Save the returned data.task_id and poll /api/v1/task/status.

Pricing

Gemini Omni Flash charges a fixed price per video, selected by resolution and duration. Reference images are not billed.

Resolution	4 s	6 s	8 s	10 s
720p	481	577	674	770
1080p	481	577	674	770
4k	1,443	1,539	1,635	1,731

Values are credits per video. The default 6-second 720p video costs 577 credits. /pricing remains the current source of truth.

See the Gemini Omni Flash API reference for the machine-readable Schema.

Getting started

Agent integrations

Video Generation

Image Generation

Audio Generation

Gemini Omni Flash