Video

Gemini Omni

Gemini Omni brings Google's multimodal reasoning into video generation. Use APIXO to create async Gemini Omni videos from text, image references, video inputs, reusable audio assets, and character asset IDs.

Start creating API Docs

720p / 1080p / 4k

4 / 6 / 8 / 10 seconds

Image, video, audio, and character references

16:9 / 9:16

Generated preview

Gemini Omni

Start now

Create with Gemini Omni

Loading workspace...

Authentication & Integration

Checking API access...

Sign in to claim free credits and create with this model.

Integrate with AI

Copy a prompt for your AI assistant to install APIXO MCP, read the live schema, and wire this model into your app.

Endpoints

POSThttps://api.apixo.ai/api/v1/generateTask/gemini-omniCreate task

GEThttps://api.apixo.ai/api/v1/statusTask/gemini-omni?taskId={taskId}Poll status

Request body

request_type

async

Use callback with callback_url for webhooks.

input

object

Mode, prompt, references, size, duration, and audio settings.

Parameters

request_typerequired

Typeselect

Use async to poll statusTask, or callback to receive a webhook when the result is ready.async · callback

callback_urlconditional

Typetext

Public HTTPS URL for callback delivery.Only in request_type=callback

moderequired

Typeselect

Gemini Omni supports audio asset, character asset, and async video generation modes. The browser Playground uses video mode.gemini-omni-audio · gemini-omni-character · gemini-omni-video

promptoptional

Typestring

Required for character and video modes. Optional voice description for audio mode.

voice_keyoptional

Typetext

Audio mode only. Official base voice key such as achernar, aoede, charon, kore, or zephyr.

nameoptional

Typetext

Required for audio asset creation and optional for character asset creation.

preview_textoptional

Typestring

Audio mode only. Optional preview dialogue for the generated voice asset.

audio_idsoptional

Typelist

Optional audio asset IDs returned by gemini-omni-audio. Video mode supports up to 3 IDs.

character_idsoptional

Typelist

Optional character asset IDs returned by gemini-omni-character. Video mode supports up to 3 IDs; each ID uses one multimodal quota unit.

resolutionoptional

Typeselect

Required for video generation. Supported values are 720p, 1080p, and 4k.720p · 1080p · 4k

durationoptional

Typeselect

Required for video generation when no source video is provided. Ignored when video_urls is present.4 · 6 · 8 · 10

aspect_ratiooptional

Typeselect

Optional output aspect ratio for video generation.16:9 · 9:16

video_startoptional

Typetext

Optional source-video start position. Requires video_urls and must be non-negative.

video_endoptional

Typetext

Optional source-video end position. Requires video_urls and must be greater than or equal to video_start.

seedoptional

Typetext

Optional non-negative random seed for video generation.

Parameter	Type	Required	Description
`request_type`	select	required	Use async to poll statusTask, or callback to receive a webhook when the result is ready.async · callback
`callback_url`	text	conditional	Public HTTPS URL for callback delivery.Only in request_type=callback
`mode`	select	required	Gemini Omni supports audio asset, character asset, and async video generation modes. The browser Playground uses video mode.gemini-omni-audio · gemini-omni-character · gemini-omni-video
`prompt`	string	optional	Required for character and video modes. Optional voice description for audio mode.
`voice_key`	text	optional	Audio mode only. Official base voice key such as achernar, aoede, charon, kore, or zephyr.
`name`	text	optional	Required for audio asset creation and optional for character asset creation.
`preview_text`	string	optional	Audio mode only. Optional preview dialogue for the generated voice asset.
`audio_ids`	list	optional	Optional audio asset IDs returned by gemini-omni-audio. Video mode supports up to 3 IDs.
`character_ids`	list	optional	Optional character asset IDs returned by gemini-omni-character. Video mode supports up to 3 IDs; each ID uses one multimodal quota unit.
`resolution`	select	optional	Required for video generation. Supported values are 720p, 1080p, and 4k.720p · 1080p · 4k
`duration`	select	optional	Required for video generation when no source video is provided. Ignored when video_urls is present.4 · 6 · 8 · 10
`aspect_ratio`	select	optional	Optional output aspect ratio for video generation.16:9 · 9:16
`video_start`	text	optional	Optional source-video start position. Requires video_urls and must be non-negative.
`video_end`	text	optional	Optional source-video end position. Requires video_urls and must be greater than or equal to video_start.
`seed`	text	optional	Optional non-negative random seed for video generation.

File parameters (URL arrays)

image_urlsCharacter mode requires exactly one image. Video mode accepts optional reference images; each image uses one multimodal quota unit.

video_urlsVideo mode accepts up to one source video. A video URL uses two multimodal quota units; when present, duration is ignored and billing is fixed per request.

Example (cURL)

curl -X POST "https://api.apixo.ai/api/v1/generateTask/gemini-omni" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "request_type": "async",
    "input": {
      "mode": "gemini-omni-video",
      "prompt": "A cinematic macro shot of a translucent glass hummingbird hovering above a rain-soaked neon flower, soft reflections, realistic motion, shallow depth of field",
      "resolution": "720p",
      "duration": "4",
      "aspect_ratio": "16:9"
    }
  }'

Full API documentation

Gemini Omni Tips

Create video generation with Gemini Omni

Use Gemini Omni through APIXO to try prompts, inspect examples, compare costs, and keep useful settings ready for future generations.

Create

Start with an idea

Pick a model, tune the settings, and create in one place.

Write an idea

Describe what you want to make.

Use an example

Start from a ready-made style.

Adjust the look

Tune the key settings quickly.

Create

Generate when everything feels right.

Keep results

Save what you want to revisit.

Try again

Make the next version faster.

Gemini Omni API Technical Specs

Current Gemini Omni capabilities and important creation limits.

Gemini Omni Modes

Gemini Omni video generation with optional multimodal references

Gemini Omni Output

720p / 1080p / 4k

Gemini Omni Controls

4 / 6 / 8 / 10 seconds / Image, video, audio, and character references

Gemini Omni Latency

Typically 1-3 minutes

Google Multimodal Video API

Key Gemini Omni capabilities

Gemini Omni video generation

Use Gemini Omni for Gemini Omni video generation with optional multimodal references creation with simple controls for prompts, references, and output settings.

Gemini Omni reference control

Upload supported reference assets where available, then keep refining the same idea on the model page.

Gemini Omni generation status

Keep track of longer video generations while the result is being prepared.

What can you build with Gemini Omni?

Gemini Omni product videos

Create product reveals, ecommerce ads, launch clips, and campaign motion assets.

Gemini Omni social creative

Generate short videos for social channels and iterate quickly across prompt, duration, resolution, and ratio settings.

Gemini Omni storyboard workflows

Use text prompts and references to validate scenes, shots, and motion ideas before committing production budget.

Read FAQ

Gemini Omni notes and questions

The browser Playground is fixed to gemini-omni-video for this launch; audio and character asset creation are documented for API use.

Video generation requires prompt and resolution. Duration is required only when video_urls is not provided.

Supported video durations without source video are 4, 6, 8, and 10 seconds.

Video mode supports up to 3 audio IDs, up to 3 character IDs, and up to 1 source video URL.

The multimodal quota from image_urls, video_urls, and character_ids cannot exceed 7 units. Each image uses 1 unit, one source video uses 2 units, and each character ID uses 1 unit.

video_start and video_end require video_urls, cannot be negative, and video_end must be greater than or equal to video_start.

Actual latency varies by prompt complexity, reference inputs, selected resolution, and provider queue load.

Yes. Use the Create tab to try prompts, upload supported inputs, and adjust settings directly in the browser.

APIXO

Create with Gemini Omni

Write a prompt, adjust the style, and generate your next image in one place.

Start creating View API docs