# Gemini Omni Flash

Google's first Gemini Omni Flash: conversational video with sound at a fixed resolution.

Last updated: 2026-10-05

**Key facts**

- Gemini Omni Flash is Google's first Gemini Omni video model, and in Sloosh it makes 3 to 10 second clips with sound in 16:9 or 9:16.
- Gemini Omni Flash has no resolution setting in Sloosh: the model picks it.
- Gemini Omni Flash takes up to 6 reference images, 1 video of up to 10 seconds and an optional start frame, but no end frame.
- Gemini Omni Flash accepts prompts of up to 50,000 characters in Sloosh.

Web page: https://sloosh.ai/docs/gemini-omni-flash

Gemini Omni Flash is the first Omni model. It makes 3 to 10 second clips with sound and picks the resolution itself. Google has moved to [Gemini Omni 1.1 Flash](https://sloosh.ai/docs/gemini-omni-1-1-flash.md), which adds resolution control and end frames, so start new work there.

## When to use it

- Re-running an older prompt that was made on it.
- For new work, use [Gemini Omni 1.1 Flash](https://sloosh.ai/docs/gemini-omni-1-1-flash.md).

## How to prompt it

- The same as 1.1: describe the scene and action plainly, with the sound on its own line.
- It takes a start frame but no end frame.

[Prompting Gemini Omni](https://sloosh.ai/docs/prompting-gemini-omni.md) has the full guide.

## Example

Text only: one action and its sounds.

[![A golden retriever puppy tumbling out of a cardboard box onto a rug](https://sloosh.ai/docs/m-gemini-omni-flash.webp)](https://sloosh.ai/docs/m-gemini-omni-flash.mp4) (video)

_16:9, 6 seconds, Enhance off. Clips here play without sound; press one to pause._

**Prompt**

```text
A golden retriever puppy tumbles out of a cardboard delivery box onto a living-room rug, shakes itself off and looks up at the camera. Low camera at puppy height, warm afternoon light.
Audio: cardboard rustling, a tiny happy bark.
```

## Settings in Sloosh

- Aspect ratio: 16:9 or 9:16.
- Length: 3 to 10 seconds.
- References: up to 6 images and 1 video of up to 10 seconds.
- Start frame: optional; no end frame.
- Sound: always on.
- Prompt: up to 50,000 characters.

Open it in [Video](https://sloosh.ai/studio/video) and pick Gemini Omni Flash in the model chip. The Generate button shows the cost first.

## Frequently asked questions

### What is the difference between Gemini Omni Flash and Gemini Omni 1.1 Flash?

Gemini Omni 1.1 Flash adds resolution control (360p to 4K), an end frame for first-and-last-frame clips, and clip extension up to 40 seconds. Both make 3 to 10 second clips with sound in Sloosh and take 6 reference images and one video, so start new work on [Gemini Omni 1.1 Flash](https://sloosh.ai/docs/gemini-omni-1-1-flash.md).

### Can I set the resolution in Gemini Omni Flash?

No. Gemini Omni Flash has only aspect ratio (16:9 or 9:16) and length settings in Sloosh, and the model picks the resolution. For 360p drafts or 1080p and 4K finals, use [Gemini Omni 1.1 Flash](https://sloosh.ai/docs/gemini-omni-1-1-flash.md).

### Can Gemini Omni Flash extend a video?

No. In Sloosh, extending a clip in 10-second steps up to 40 seconds is a Gemini Omni 1.1 Flash feature only. With Gemini Omni Flash, make several clips and join them with Video Concat in [Spacelab](https://sloosh.ai/docs/spacelab.md).

### Can I turn off sound in Gemini Omni Flash?

No. Gemini Omni Flash always makes sound in Sloosh, and there is no switch to turn it off. Give the sound its own Audio line in the prompt, such as "Audio: cardboard rustling, a tiny happy bark.", because without one the model guesses.

## Sources

- [Gemini API: Omni](https://ai.google.dev/gemini-api/docs/omni)
- [Gemini API: changelog](https://ai.google.dev/gemini-api/docs/changelog)