What is Gemini Omni 1.2 Flash Reference-to-Video?

Gemini Omni 1.2 Flash Reference-to-Video generates videos from a text prompt with image or video references. Provide at least one reference image or reference video on each::labs to guide the result, then select an output duration, resolution, and aspect ratio.

When should I choose Gemini Omni 1.2 Flash Reference-to-Video?

Choose Gemini Omni 1.2 Flash Reference-to-Video when your video should follow supplied visual references. It supports short video workflows for creators who want to describe a scene with a prompt and guide generation with image or video material rather than text alone.

Which output options does Gemini Omni 1.2 Flash support?

Gemini Omni 1.2 Flash supports generated video durations from 3 to 15 seconds. Select 360p, 720p, 1080p, or 4k resolution and a landscape or portrait aspect ratio. Billing follows token usage, so reference media and reasoning can add to the estimated video output charge.

Google | Gemini Omni 1.2 Flash | Reference to Video.

Google
Text to Video
$0.0232–$0.2085 per second, depending on resolution

Gemini Omni 1.2 Flash Reference-to-Video generates videos from image or video references, with 3-15 second durations and resolution controls for creators.

View API docs

Request

Prompt *

Create a 15-second horizontal UGC-style video featuring the red-haired woman from the reference image, sitting comfortably on the edge of a bed in a cozy, modern bedroom filled with soft morning sunlight.

She speaks directly to the camera in a relaxed, playful, and authentic tone, as if showing her latest fashion find to a friend.

The product is the hat from the reference image, featuring the exact embroidered text "SHIP HAPPENS". Preserve its original color, shape, fabric, typography, and design. The text must remain perfectly accurate, clearly readable, and unchanged throughout the video.

She holds the hat naturally, turns it slightly to show the embroidery, and brings it closer to the camera so the words "SHIP HAPPENS" are clearly visible. She then puts it on, gently adjusts the brim, and smiles.

Preserve her exact appearance and outfit. Only one hat appears throughout the video.

Dialogue (natural American English, playful and conversational):

"Okay, can we talk about this hat for a second? It literally says 'Ship Happens,' and I'm obsessed. It's funny, it's cute, and honestly, the fit is perfect. I've been wearing it with everything lately. I mean, who doesn't love a good pun? You seriously need this hat!"

Duration

Aspect Ratio

Resolution

00:00

What Google | Gemini Omni 1.2 Flash | Reference to Video is good at.

Related model
PixVerse V6 Max Text-to-Video
Run it on the same input to compare output, speed and price.
View model

Model documentation

About this model

Gemini Omni 1.2 Flash Reference-to-Video generates videos from image or video references, with 3-15 second durations and resolution controls for creators.

Questions about this model

The short answers. For anything specific to your setup, talk to our team.

Talk to an engineerPlan for your expected volume.

Gemini Omni 1.2 Flash Reference-to-Video generates videos from a text prompt with image or video references. Provide at least one reference image or reference video on each::labs to guide the result, then select an output duration, resolution, and aspect ratio.