# Gemini Omni 1.2 Flash

Gemini Omni 1.2 Flash is Google's newest Omni video model, released on 5 November 2026. It makes and edits video with sound. In Sloosh it makes 3 to 15 second clips from 360p up to 4K in 16:9 or 9:16, from text, up to 6 reference images, a reference video, or first and last frames.

Web page: https://sloosh.ai/models/gemini-omni-1.2-flash

[Use Gemini Omni 1.2 Flash](https://sloosh.ai/studio/video?video-model=gemini-omni-1.2-flash)

## Key facts in Sloosh

|   |   |
| --- | --- |
| Made by | Google |
| Makes | video (with sound: always) |
| Inputs | text, start frame (optional), end frame (optional, needs a start frame), up to 6 reference images, 1 reference video (up to 10 seconds), or a video to edit (up to 15 seconds) |
| Length | 3, 5, 8, 10 or 15 seconds; up to 10 seconds with an end frame |
| Resolution | 360p, 720p, 1080p, 4K |
| Aspect ratios | 16:9, 9:16 |
| Extend | 10-second steps in Spacelab, up to 30 seconds in total |
| Prompt limit | 50,000 characters |

## Best for

- A full scene in one clip. Google raised the limit to 15 seconds, up from 10 on 1.1 ([Google](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/)).
- Changing footage you already have, of up to 15 seconds, such as swapping a product or restyling a scene. Google advises adding "Keep everything else the same" ([Google](https://aistudio.google.com/docs/omni)).
- Product and fashion shots with a moving camera. Google says faces, garments, packaging type and brand colours stay consistent across camera angles, with fewer unwanted jump cuts than 1.1 ([Google](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/)).
- People talking on screen. Google says 1.2 brings more natural dialogue, clearer lip-sync and distinct voices for several speakers ([Google Cloud](https://cloud.google.com/blog/products/ai-machine-learning/gemini-omni-1-2-flash-generally-available)).
- Transitions and loops between two images, in clips of up to 10 seconds.
- Drafting. Google suggests 360p or 720p for drafts and 1080p or 4K for finals ([Google](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/)).
- Product placement in existing clips.

## How to use Gemini Omni 1.2 Flash in Sloosh

1. Open [Video in Studio](https://sloosh.ai/studio/video?video-model=gemini-omni-1.2-flash) with Gemini Omni 1.2 Flash in the model chip.
2. Add what you have: reference images (up to 6), a start frame and optional end frame, or one reference video of up to 10 seconds. To change a clip you already have, of up to 15 seconds, open Edit in Studio instead.
3. Pick 16:9 or 9:16, the length (3, 5, 8, 10 or 15 seconds; up to 10 with an end frame) and the resolution. Try 360p first.
4. Write one moment as one continuous shot, with its own Audio line.
5. Press Generate. The price shows on the button first. For longer ads, extend the clip in 10-second steps with Spacelab's Video edit node, up to 30 seconds in total.

## Gemini Omni 1.2 Flash prompting tips

- Layer the prompt: shot framing and motion, style, lighting, location, action ([Google DeepMind](https://deepmind.google/models/gemini-omni/prompt-guide/)).
- Say "one continuous shot, no cuts" if you want a single scene, or the model may add shots ([Google](https://aistudio.google.com/docs/omni)).
- Time events with natural language or timecodes like `[0-5s]` ([Google](https://aistudio.google.com/docs/omni)).
- With first and last frames, describe only the change between them ([Google DeepMind](https://deepmind.google/models/gemini-omni/prompt-guide/)).
- Keep edit prompts short. Google says overly detailed ones can cause changes you did not ask for ([Google](https://aistudio.google.com/docs/omni)).
- For clips over 10 seconds, keep a reference video to 3 seconds or less, which is Google's limit at those lengths ([Google](https://aistudio.google.com/docs/omni)).
- When you extend, name anyone who left the frame earlier. Google says the model reads only the last 10 seconds of the clip for context ([Google](https://aistudio.google.com/docs/omni)).
- Describe the audio, including music, in the prompt ([Google](https://aistudio.google.com/docs/omni)).
- For on-screen text, write the exact words; Sloosh's guide says keep to one caption per clip ([Sloosh docs](https://sloosh.ai/docs/prompting-gemini-omni.md)).

## Gemini Omni 1.2 Flash example prompts

**Prompt 1**

```text
One continuous 15-second shot, no cuts. Photoreal. A small bakery at dawn, warm light from the ovens. [0-5s] The camera glides in through the front door, past shelves of bread. [5-11s] It follows a baker as she slides a tray of croissants onto the counter. [11-15s] She looks up, smiles and says, "First batch of the day." Audio: oven fans humming, the tray landing on steel, her voice close and clear.
```

**Prompt 2**

```text
A single continuous 15-second shot. Photoreal. A sunny kitchen counter, morning light through the window behind. The camera pushes in slowly on the blender from the first image as a hand pours a thick pink smoothie into a glass, then lifts the glass toward the window. Audio: the blender winding down, the pour, soft acoustic guitar.
```

**Prompt 3**

```text
Replace the bottle in the woman's hand with the bottle in the first image. Keep everything else the same.
```

## How it compares in Sloosh

|   | Gemini Omni 1.2 Flash | Gemini Omni 1.1 Flash | Kling V3 Omni |
| --- | --- | --- | --- |
| Length | 3, 5, 8, 10 or 15 s | 3 to 10 s | 3 to 15 s |
| References | 6 images, 1 video | 6 images, 1 video | 7 images |
| Resolution | 360p to 4K | 360p to 4K | 720p, 1080p |
| Ratios | 16:9, 9:16 | 16:9, 9:16 | 16:9, 9:16, 1:1 |
| Sound | always | always | optional |

## Where else it runs

- The Gemini API, Google AI Studio, Google Cloud, Google Flow, Google Vids and Google Ads, plus the Gemini app, where 15-second videos are rolling out to Google AI Ultra subscribers ([Google](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/), 5 Nov 2026).

## Frequently asked questions

### What is Gemini Omni?

Google's video model family that generates and edits video with sound from text, images and video. Gemini Omni 1.2 Flash is the newest, released on 5 November 2026. Sloosh offers it beside [Gemini Omni 1.1 Flash](https://sloosh.ai/models/gemini-omni-1.1-flash).

### How long can a Gemini Omni 1.2 Flash video be?

3, 5, 8, 10 or 15 seconds per clip in Sloosh, and up to 10 seconds with an end frame. For longer videos, extend a clip in Spacelab in 10-second steps, up to 30 seconds in total.

### What is the difference between Gemini Omni 1.2 Flash and 1.1 Flash?

In Sloosh, 1.2 makes clips of up to 15 seconds and edits videos of up to 15 seconds; [Gemini Omni 1.1 Flash](https://sloosh.ai/models/gemini-omni-1.1-flash) stops at 10 seconds for both. Extending in Spacelab, 1.1 reaches 40 seconds in total and 1.2 reaches 30. Google says 1.2 also holds references better and makes fewer unwanted jump cuts ([Google](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/)).

### Can Gemini Omni make 4K video?

Yes. Sloosh offers 360p, 720p, 1080p and 4K. Google says 1080p and 4K are upscaled ([Google](https://aistudio.google.com/docs/omni)).

### Does Gemini Omni use the audio from my video?

No. Google says any audio in a video reference is ignored ([Google](https://aistudio.google.com/docs/omni)). Write the sound you want on an Audio line.

### How do I use Gemini Omni?

Open Video in Studio, pick Gemini Omni 1.2 Flash, add images, frames or a video, write one continuous shot with an Audio line and press Generate.

### Is Gemini Omni free?

No. Sloosh has no free plan. Studio shows the exact price before you generate, and [pricing](https://sloosh.ai/pricing) has the plans. Your first Sloosh workspace gets starter credits once, so you can try it before you pay.

### Which aspect ratios does Gemini Omni support?

16:9 and 9:16, in Sloosh and in Google's API.

## Related

- Models: [Gemini Omni 1.1 Flash](https://sloosh.ai/models/gemini-omni-1.1-flash), [Kling V3 Omni](https://sloosh.ai/models/kling-v3-omni-video), [Kling 4.0](https://sloosh.ai/models/kling-4.0)
- Spaces: [Product Placement Video](https://sloosh.ai/spaces/product-placement-video), [Swap Product into a Reference Ad](https://sloosh.ai/spaces/swap-product-into-a-reference-ad)
- Docs: [Gemini Omni 1.2 Flash](https://sloosh.ai/docs/gemini-omni-1-2-flash.md), [Prompting Gemini Omni](https://sloosh.ai/docs/prompting-gemini-omni.md)

## Sources

- [Google: Gemini Omni 1.2 Flash, fast and cost-effective with longer clips](https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemini-omni-1-2-flash/)
- [Google Cloud: Gemini Omni 1.2 Flash now GA](https://cloud.google.com/blog/products/ai-machine-learning/gemini-omni-1-2-flash-generally-available)
- [Gemini API: Omni](https://aistudio.google.com/docs/omni)
- [Google DeepMind: Gemini Omni prompt guide](https://deepmind.google/models/gemini-omni/prompt-guide/)
- [Sloosh docs: Gemini Omni 1.2 Flash](https://sloosh.ai/docs/gemini-omni-1-2-flash)
- [Sloosh docs: Prompting Gemini Omni](https://sloosh.ai/docs/prompting-gemini-omni)