muapi Logomuapi Logo
muapi Logomuapi Logo

Gemini Omni 1.2 Flash Reference to Video: Image-to-Video

Generate video guided by up to 6 reference images with Gemini Omni 1.2 Flash. 3-15s clips with native audio, up to 4K. Try free. Gemini Omni 1.2 Flash — generate a 3-15 second video guided by up to 6 reference images for subject, scene, or style, with native synchronized audio.

Interactive model controls

Gemini Omni 1.2 Flash — generate a 3-15 second video guided by up to 6 reference images for subject, scene, or style, with native synchronized audio.

📝

Overview

About this model

Gemini Omni 1.2 Flash Reference to Video generates a 3-15 second video guided by 1-6 reference images that define subjects, scenes, style, or a storyboard. Unlike image-to-video, the images are conditioning references rather than a literal first frame, so the model is free to compose new motion and camera work around them, and it returns natively synchronized audio in the same pass. Select 16:9, 9:16, or auto aspect ratio and render from 360p to 4K. Pricing is per second of output ($0.023/s at 360p, $0.07/s at 720p, $0.105/s at 1080p, $0.21/s at 4K). For a literal starting frame use Gemini Omni 1.2 Flash Image to Video; other 1.2 Flash endpoints: Gemini Omni 1.2 Flash, Gemini Omni 1.2 Flash Image to Video, Gemini Omni 1.2 Flash Video Edit.

1Character consistency: Keep the same character or product across new scenes.
2Style transfer: Use reference art to set the look of a generated clip.
3Storyboard-to-video: Feed storyboard panels as references to guide the scene.
4Brand content: Place a product from reference shots into new environments.
5Rapid iteration: Draft at 360p before upscaling the final render.
💰

Pricing & Value

Cost analysis

muapi$0.023–$0.21 per second (by resolution)

Pay-per-generation, no subscription. Billed per second of generated video.

Fal.aiNot available

This model is not listed on Fal.ai.

ReplicateNot available

This model is not listed on Replicate.

** Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Promptstring

Describe the scene, action, camera movement, and sound.

Default ValueThe character from the first image walks through the city in the second image at sunset, ambient street sounds
Reference Imagesarray

1-6 reference images used for subject, scene, style, or storyboard reference (not a starting keyframe).

Default Value-
Durationinteger

Video length in seconds (3-15).

Default Value8
Aspect RatioEnum (3 options)

Output aspect ratio. `auto` follows the reference image and is treated as 16:9 when none is provided.

Default Value16:9
ResolutionEnum (4 options)

Output resolution. Output priced per second of generated video: $0.023/s at 360p, $0.07/s at 720p, $0.105/s at 1080p, $0.21/s at 4K.

Default Value720p
Seedinteger

Random seed (10000-99999). Omit for a random seed.

Default Value-
📖

Implementation Guide

Developer documentation

How to Use Gemini Omni 1.2 Flash Reference to Video

  1. Collect reference images Provide 1-6 publicly accessible image URLs in image_urls (each up to 20MB).

  2. Write the prompt Describe how the references should appear: 'The character from the first image walks through the city in the second image at sunset.'

  3. Pick duration, aspect ratio, and resolution Duration 3-15s. Billing is $0.023/s at 360p, $0.07/s at 720p, $0.105/s at 1080p, $0.21/s at 4K.

  4. Submit and poll

curl -X POST https://api.muapi.ai/api/v1/gemini-omni-flash-1-2-reference-to-video
-H "Authorization: Bearer YOUR_API_KEY"
-H "Content-Type: application/json"
-d '{"prompt": "The character from the first image walks through the city in the second image at sunset", "image_urls": ["https://example.com/ref1.png", "https://example.com/ref2.png"], "duration": 8, "resolution": "720p"}'

Then poll `GET /api/v1/predictions/{request_id}/result` until `status` is `completed`.
❓

Common Questions

Frequently asked

What is Gemini Omni 1.2 Flash Reference to Video?

It generates a video conditioned on 1-6 reference images for subject, scene, or style, with native audio.

How many reference images can I use?

Between 1 and 6, each up to 20MB.

How is it priced?

Per second of generated video: $0.023/s at 360p, $0.07/s at 720p, $0.105/s at 1080p, $0.21/s at 4K.

Are reference images used as the first frame?

No. They are conditioning references. Use [Image to Video](/playground/gemini-omni-flash-1-2-image-to-video) for a literal first frame.

What durations are supported?

3-15 seconds.

Does it generate audio?

Yes, dialogue, ambient sound, and music are generated natively.