Gemini Omni Flash 1.2 AI video generator

Gemini Omni Flash 1.2 AI video generator on Morphic

Google's fifteen-second video model. Type the shot, then direct it.

Every video model

One workspace

Shot the Gemini Omni Flash 1.2 way

Key features

What Gemini Omni Flash 1.2 generates

A whole beat, sound included

Fifteen seconds is long enough for an approach, a turn, and an arrival, with the room tone, the effects, and the spoken line arriving in the same pass as the picture.

A night fish market under hanging lamps, steam rising as a vendor wraps an order

Edits you type

Generate, then direct: relight the scene, swap the wardrobe, rewrite the sign. Each instruction lands on the same take while everything unmentioned stays exactly as shot.

Split frame of the same man on a stone bridge, charcoal coat on the left and olive raincoat on the right

One master, both ratios

Approve the landscape cut and ask for the vertical one as a follow-up, so the social version is the same scene rather than a second roll of the dice.

A tram-stop shot shown as a wide landscape frame beside a vertical crop of the same moment

Reference-true people and products

Six images pin who and what is in the shot, three of them for each of two characters, so a face, an outfit, and a label survive from the first take to the last edit.

Reference cards of an invented presenter and a ribbed amber bottle pinned beside the finished campaign frame

Generate a video in three steps

  1. 01

    Open Morphic

    Sign up and start generating on a free-flowing, infinite visual canvas.

  2. 02

    Write the shot with its sound

    Subject, action, camera, light, and the audio. Say what happens when, and whether it plays as one continuous shot or cuts.

  3. 03

    Generate, then direct

    Refine the take in follow-ups: one change per turn, and extend it when the scene needs room.

All on Morphic

Your complete video stack

Simple pricing

Get started for free today, with the option to upgrade or cancel anytime.

Basic

$19/ month
billed as $228 per year

2400 monthly credits

1 user only

All models

Workflows

Standard

$24/ month
billed as $288 per year

3625 monthly credits

1 user only

All models

Workflows

Pro

$45/ month
billed as $540 per year

6350 shared monthly credits

1 user

+ up to 4 more at extra cost

All models

Workflows

Max

$170/ month
billed as $2040 per year

24650 shared monthly credits

1 user

+ up to 9 more at extra cost

All models

Workflows

Enterprise

For higher limits

Custom

pricing and billing terms

High-volume credits
Custom seat limits
All models
Workflows
Pricing Gradient

Free

For playing around

$0

forever free

Up to 20 credits
1 user only
Limited models
Workflows

FAQs

What is Gemini Omni Flash 1.2?
The third release of Google DeepMind's multimodal video model, written by Google as Gemini Omni 1.2 Flash and announced on 5 November 2026. It generates 3 to 15 second clips with native audio from text, images, and short reference clips, then keeps working on them: follow-up prompts edit the same take across its whole length, and extensions grow it to 30 seconds. Version 1.2 added the fifteen-second ceiling, whole-clip editing, and a per-second price about 30% below 1.1.
Is Gemini Omni Flash 1.2 on Morphic?
Not yet. Its predecessor, Gemini Omni Flash 1.1, generates on Morphic today with native audio, alongside Veo 3.1, Seedance 2.5, Kling, and the rest of the video catalogue. Every example on this page was made on Morphic with the closest live model, not with 1.2. Morphic has a free plan, so you can open the workspace and try the wider catalogue first.
Can I generate a video with native sound right now?
Yes. Type the shot into the prompt bar above and generate, with everything running in the browser and nothing to install. Write the audio into the brief the way you write the picture, because a prompt that says nothing about sound comes back with a track the model chose. Even silence is something you ask for.
What makes this different from a standard text-to-video tool?
The generation is a conversation rather than a one-shot render. After the first take you say what to change, in plain language, and it lands on the same footage while the rest of the frame stays as shot. A standard generator makes you re-roll the whole clip and hope; this keeps the take and fixes the detail.
How long can a generated video be?
Gemini Omni Flash 1.2 generates 3 to 15 whole seconds in one pass, with ten as the default, where 1.1 stopped at ten. From there, extension continues the clip in turns of three to ten seconds up to 30 seconds in total, reading the final ten seconds so motion, characters, and audio carry across the join.
Can I get the vertical version of a landscape clip?
In 1.2, yes, as a follow-up: an approved 16:9 take can be asked for 9:16 rather than generated again from the prompt. On Morphic today, generate the ratio you need from the picker and keep the same brief, so the two cuts stay recognisably the same scene.
Can it work from my own images and clips?
Yes. Animate a single still, give it an opening and closing frame to interpolate between, or attach reference images that pin a character, a product, or a palette. Gemini Omni Flash 1.2 takes up to six reference images, three of them each for up to two characters, plus motion clips of up to three seconds.
How do I generate video online without installing anything?
Everything runs in the browser on Morphic, with nothing to install and the rendering handled server side. Type the prompt, generate, and the render lands on your Canvas from any laptop. From there you can refine the take in follow-ups, run the same brief through another model, and compare the results side by side.

You might also like