---
title: "Gemini Omni Flash 1.2 AI Video Generator with Sound | Made"
description: "Generate 3 to 15 second videos with sound and spoken lines using Google Gemini Omni Flash 1.2 in Made, from a prompt, a start image, or start and end frames. See original examples, formats, strengths, and limits. On Made: from 0.27 credits / second."
url: https://www.made.app/models/gemini-omni-flash-1-2
---

[Models](https://www.made.app/models) / Gemini Omni Flash 1.2

Google · video model

# Gemini Omni Flash 1.2 Generate AI video

from 0.27 credits / second · no subscription · [price breakdown](https://www.made.app/price-sheet/gemini-omni-flash-1-2) · [API playground](https://www.made.app/playground/gemini-omni-flash-1-2)

Prompt

4s

[Start from your own image →](https://www.made.app/tools/ai-video-generator?model=gemini-omni-flash-1.2)

Made with Gemini Omni Flash 1.2 · Action, sound, and a spoken line

Gemini Omni Flash 1.2 is the newest version of Google’s multimodal video model. It makes 3 to 15 second clips with native sound from a text prompt, a start image, or a start and an end frame, at 360p, 720p, 1080p, or 4K, and the people on screen say the lines you write.

Original results

## Gemini Omni Flash 1.2 examples made in Made

These examples use media produced for Made. Prompts are included when the production record allows them.

### Action, sound, and a spoken line

“Handheld phone video in a laundromat late at night under flickering fluorescent lights: a guy in socks sits in a rolling laundry cart and his friend shoves it hard down the aisle between the dryers. The cart rattles, spins, and coasts into a mountain of clean towels on a folding table, which collapses on top of him. The friend filming wheezes with laughter and says, "Again. Do it again." Dryers tumbling, the cart wheels squeaking, laughter.”

### A photo that talks and moves

“She looks straight into the lens and says, "Forty years. Never missed a strike." Then she turns, steps up to the lane and rolls the ball. The pins crash, and she throws both arms in the air while her friends cheer behind her. Handheld phone camera, bowling alley noise, the rumble of the ball.”

### 15 second scene with cuts

“A 15 second scene in three shots. Shot 1, wide: early morning in a city park, an old man on a bench tosses breadcrumbs to a crowd of pigeons; one pigeon is wearing a tiny canvas backpack. Shot 2, close-up at ground level: the backpack pigeon tugs the little zipper open with its beak and pulls out a folded paper note. Shot 3, medium shot: the old man unfolds the note, reads it, shakes his head, and laughs as he says, "She still owes me five bucks." Soft morning light, pigeons cooing, distant traffic, the rustle of paper.”

![Direct-flash photo of a grandmother in a cardigan holding a bowling ball up to the camera in a bowling alley at night, the start frame for a Gemini Omni Flash 1.2 clip](https://www.made.app/images/models/gemini-omni-flash-1-2/still-01.jpg)

### Start-frame control

“Use this image as the opening composition.”

![A frame from the Gemini Omni Flash 1.2 laundromat clip, the cart rolling between the dryers](https://www.made.app/images/models/gemini-omni-flash-1-2/still-02.jpg)

### Frame from the action clip

“A single frame from the clip above, straight from Gemini Omni Flash 1.2.”

![A frame from the Gemini Omni Flash 1.2 park clip, a pigeon with a tiny backpack among the flock](https://www.made.app/images/models/gemini-omni-flash-1-2/still-03.jpg)

### Frame from the 15 second scene

## Where Gemini Omni Flash 1.2 fits best

### Best uses

- Talking clips: write a line in quotes and the person on screen says it
- Scenes with cuts, up to 15 seconds in one generation
- Bringing a photo to life, or moving between a start and an end frame
- Cheap drafts at 360p, then the keeper at 720p, 1080p, or 4K

### Limits to know

- Aspect ratios are 16:9 and 9:16 only. Image-to-video crops the start image to the shape you pick.
- With an end frame, clips run at most 10 seconds.
- Sound is always on. There is no silent mode, and no negative prompt or seed.

## Specifications in Made

These are the controls exposed by Made, which may be narrower than a provider’s full API.

**Aspect ratios**

16:9, 9:16

**Resolution**

360p (default in Made), 720p, 1080p, 4K

**Duration**

3–15 seconds (up to 10 with an end frame)

**Output formats**

MP4

**Reference images**

2

**Processing**

Under a minute for most clips in our tests, about a minute for 15 seconds at 720p; progress appears in the generation view.

## How to use Gemini Omni Flash 1.2 in Made

1. **Choose text or image input**
   Write a prompt on its own, or add a start image, and optionally an end frame, to anchor the clip.
2. **Direct the action and the sound**
   Say what moves, how the camera follows it, and what we hear. Put dialogue in quotes and the person on screen says it.
3. **Set format and length**
   Pick 16:9 or 9:16 and any length from 3 to 15 seconds, or up to 10 with an end frame.
4. **Draft at 360p, finish higher**
   Try ideas at 360p, then run the take you like at 720p, 1080p, or 4K.

Capabilities reviewed 2026-11-05 · Provider: [Google](https://fal.ai/models/google/gemini-omni-flash/v1.2/text-to-video)

Made shows the price in credits before generation. [View Made pricing](https://www.made.app/pricing)

## Go deeper on Gemini Omni Flash 1.2

### [Gemini Omni Flash 1.2 pricing](https://www.made.app/price-sheet/gemini-omni-flash-1-2)

from 0.27 credits / second — the full breakdown, verified.

### [Gemini Omni Flash 1.2 prompts](https://www.made.app/models/gemini-omni-flash-1-2/prompts)

6 tested prompts with the outputs they produced.

## Compare relevant models

video

### [Gemini Omni Flash 1.1](https://www.made.app/models/gemini-omni-flash-1-1)

Gemini Omni Flash 1.1 is the second version of Google’s fast multimodal video model, generating 3 to 10 second clips from text or a start image with sharper motion and stronger prompt adherence than v1. Gemini Omni Flash 1.2 is newer.

video

### [Kling 4 Flash](https://www.made.app/models/kling-4-flash)

Kling 4 Flash is the fast, lower-cost tier of Kling AI’s fourth-generation video model. It makes 3 to 20 second clips with native sound at 720p, from a text prompt or from a start image, and the people on screen speak the lines you write.

video

### [Seedance 2.5](https://www.made.app/models/seedance-2-5)

Seedance 2.5 is ByteDance’s newest video model in Made, generating clips up to 30 seconds in a single pass with native audio at 480p to 4K — the longest single-shot ceiling in the Made library.

## Continue with a related tool

### [AI Video Generator](https://www.made.app/ai-video-generator)

Turn any image into a captivating video. Upload a photo, describe the motion you want, and watch AI bring it to life with realistic movement, cinematic camera work, and professional quality.

### [AI Image Generator](https://www.made.app/ai-image-generator)

Create scroll-stopping product photos and marketing visuals in seconds. Describe your shot, pick your style, and get professional images ready for your store, social feed, or ad campaign.

Questions, answered

## Gemini Omni Flash 1.2 FAQ

Need a specific answer? [Contact support](mailto:support@made.app)

### What is Gemini Omni Flash 1.2?

It is the newest version of Gemini Omni Flash, Google’s multimodal video model, and the newest Gemini Omni in Made.

### How is it different from Gemini Omni Flash 1.1?

Clips run up to 15 seconds instead of 10, and every resolution costs about 10% less per second. Both make sound, take a start and an end frame, and render at 360p, 720p, 1080p, or 4K.

### Does it generate sound?

Yes. Every clip comes with native audio, including lines of dialogue you write in quotes, and there is no silent mode.

### Can I animate an existing image?

Yes. Add a start image and describe what happens next. Add an end frame too and the clip moves from one to the other, in up to 10 seconds. The image is cropped to 16:9 or 9:16, whichever you pick.

### How long can clips be?

Any whole number of seconds from 3 to 15 in one generation, or up to 10 with an end frame. The default is 4.

### Can Gemini Omni Flash 1.2 use reference images and videos, or edit a video I already have?

Yes, through the Made API and the Made MCP. Its reference model keeps the people, products, or places from up to 10 images and 3 videos of up to 15 seconds each, saved characters and products included. Its edit model changes a clip of up to 15 seconds from a written instruction, such as turning a backyard into a snowy afternoon, and the result runs about as long as the clip.

### When should I choose another model?

Choose Kling 4 Flash for 1:1 and 21:9 or clips up to 20 seconds, Kling 4 for clips up to 30 seconds, or Seedance 2.5 for photo references of people, places, and props.

---

Markdown version of https://www.made.app/models/gemini-omni-flash-1-2. Every public Made page has one at its URL plus .md (https://www.made.app/models/gemini-omni-flash-1-2.md). Site summary for AI assistants: https://www.made.app/llms.txt
