Live9 variants

Gemini Omni API

Gemini Omni is Google's natively multimodal any-to-any video family. Text, images, video, voice profiles, and character profiles can work together in a single generation flow.Muapi exposes Gemini Omni and Gemini Omni 1.1 Flash through one asynchronous API, including synchronized dialogue, ambient sound, music, video editing, reusable voices, and consistent characters.

Get a Gemini Omni API Key Try Text to Video Try Video Edit

Gemini Omni FlashNative multimodal video, audio, voice, and character workflows9 models

T2VLive

Flash

Gemini Omni Text to Video

Generate video from a text prompt with synchronized dialogue, ambient sound, and music in the same pass.

360p–4K

From $0.16

Try Model

I2VLive

Flash

Gemini Omni Image to Video

Animate up to five reference images into a scene with native synchronized audio.

360p–4K

From $0.16

Try Model

V2VLive

Flash

Gemini Omni Video Edit

Edit, restyle, relight, swap, or rewrite dialogue in source footage while preserving its motion and timing.

Up to 4K

From $2.40

Try Model

AudioLive

Flash

Gemini Omni Audio Profile

Create a reusable voice profile from one of 30 preset voices and receive an audio ID for later generations.

Voice profileFast

Free

Try Model

CharacterLive

Flash

Gemini Omni Character Profile

Create a reusable character profile from one reference image and receive a character ID.

Character profileFast

Free

Try Model

T2VLive

Flash

Gemini Omni 1.1 Flash Text to Video

Combine text with reference images, a short reference video, voice profiles, or character profiles and native audio.

Up to 4K

From $0.10/sec

Try Model

I2VLive

Flash

Gemini Omni 1.1 Flash Image to Video

Generate from a starting keyframe with an optional ending keyframe for controlled transitions.

Up to 4K

From $0.10/sec

Try Model

V2VLive

Flash

Gemini Omni 1.1 Flash Video Edit

Apply a text instruction to restyle, relight, transform, or rewrite a source video.

360p–4K

From $0.048/sec

Try Model

I2VLive

Flash

Gemini Omni 1.1 Flash Reference to Video

Guide a scene with up to seven images and three short reference-video clips.

360p–4K

From $0.048/sec

Try Model

What is Gemini Omni?

Gemini Omni is Google's native multimodal any-to-any video model family. Rather than generating video first and adding sound later, it can reason over text, images, video, and audio together while producing synchronized dialogue, ambient sound, and music.

Muapi provides the original Gemini Omni endpoints plus Gemini Omni 1.1 Flash for keyframe control, short video references, and 4K delivery. Each workflow uses the same submit-and-poll REST contract.

Gemini Omni API capabilities

Any-to-any generation

Move between text, images, video, and audio in one multimodal workflow instead of stitching separate models together.

Native synchronized audio

Generate dialogue, ambient sound, and music with the video in a shared forward pass.

T2V, I2V, and video edit

Create new scenes, animate reference images, or transform existing footage with focused endpoints.

Reusable voice profiles

Select from 30 preset voices, save the returned audio ID, and reuse a consistent voice in later calls.

Consistent character profiles

Create a character ID from one reference image and reuse it across generated scenes.

Drop-in production API

Use asynchronous Muapi requests, standard prediction polling, and pay-per-generation access without a subscription.

Gemini Omni use cases

Narrative shorts

Direct a scene, camera, character, and dialogue from one rich prompt with native audio.

Product explainers

Combine product imagery, voice profiles, and controlled video edits for clear demonstrations.

Character-led series

Reuse character and voice profiles to keep recurring talent consistent across episodes.

Source-video transformations

Restyle, relight, swap, or rewrite dialogue in existing footage with text instructions.

Reference-driven scenes

Use images and short video references to guide wardrobe, composition, motion, and tone.

Veo or Sora alternatives

Add native audio and multimodal inputs through one API surface when a general video endpoint is not enough.

Gemini Omni workflows

Workflow	Inputs	Output	Starting price
Text to Video	Prompt, optional refs, voice, character	Video + synchronized audio	From $0.16
Image to Video	Up to 5 reference images	Video + synchronized audio	From $0.16
Video Edit	Source video + instruction	Edited video	From $2.40
Audio Profile	Preset voice selection	Reusable audio ID	Free
Character Profile	One reference image	Reusable character ID	Free
1.1 Flash Reference	Up to 7 images + 3 short clips	Reference-guided video	From $0.048/sec

Quick start — Gemini Omni API

Submit a multimodal prompt to the matching Gemini Omni endpoint, save the request ID, and poll the standard result endpoint for video and audio outputs.

Submit a Gemini Omni task

curl -X POST https://api.muapi.ai/api/v1/gemini-omni-text-to-video \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A chef explains a recipe in a warm studio kitchen","resolution":"1080p","duration":8,"aspect_ratio":"16:9"}'

# Response: {"request_id":"REQUEST_ID"}

Start with the text-to-video endpoint, then add reference inputs, voice IDs, or character IDs as your workflow requires.

Retrieve video and audio

curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \
  -H "x-api-key: YOUR_API_KEY"

# The completed response contains the generated video and synchronized audio.

Poll until completion and read the generated video plus synchronized audio URL from the response.

Gemini Omni API FAQ

What is Gemini Omni API?

Gemini Omni is a natively multimodal any-to-any video family. Muapi exposes text-to-video, image-to-video, and video-edit workflows that can generate synchronized audio in the same pass.

How is Gemini Omni different from Veo or Sora?

Gemini Omni is designed around text, image, audio, and video inputs in one workflow, with native audio and focused source-video editing endpoints. The best choice depends on the controls and style your project needs.

How much does Gemini Omni cost?

Text-to-video and image-to-video start at $0.16 per generation. Video edit starts at $2.40, while 1.1 Flash video workflows are priced from $0.048 or $0.10 per second depending on the endpoint and output option.

What inputs and outputs are supported?

Use a text prompt, up to five reference images for image-to-video, or source video plus an instruction for editing. Generated video can include synchronized audio, and calls can use up to three saved voice IDs and three character IDs.

What is Gemini Omni Audio?

Gemini Omni Audio creates a reusable voice profile from 30 preset voices. Muapi returns an audio ID that can be passed to later video generations, with up to three audio IDs per request.

What is Gemini Omni Character?

Gemini Omni Character creates a reusable character profile from one reference image. The returned character ID can be used in later generations, with up to three character IDs per request.

Can Gemini Omni create vertical or square video?

Yes. Use the aspect ratio supported by the selected endpoint, including 16:9 and 9:16 compositions for landscape and vertical delivery.

Do I need a Gemini subscription?

No. Gemini Omni is available through Muapi's pay-per-generation API. Create a Muapi API key and call the endpoint directly.

Explore Gemini Omni 1.1 →Add keyframes, video references, and 4K delivery.Browse video models →Compare Gemini Omni with other live video APIs.Create an API key →Start calling Gemini Omni through Muapi.

Ready to build with Gemini Omni?

Use one asynchronous API for multimodal video, native audio, voice profiles, and consistent characters.

Get an API Key Try Gemini Omni