Gemini Omni 1.1

Googlevideolive~53s

Text-to-video with synchronized native audio, up to 4K.

from$0.042/ second

Exact price on the Run button. Failed runs cost $0.00.

How this model is billed

Reserved, then settled to the cent. A failed run costs $0.00.

Typical run ~53s: the measured mean, end to end.

Compare every price

Run Gemini Omni 1.1

Overview

Gemini Omni 1.1 is Google's text-to-video model with synchronized native audio. You describe a scene, the camera and what it should sound like in one prompt, and it returns a finished clip with the sound already in it, so there is no separate audio pass to line up afterwards.

It also works from images. Give it a first frame and the video starts exactly from that picture; add a last frame and it interpolates between the two. Or hand it a set of reference images and it lifts their subjects and styles into a brand-new scene instead of reusing them as literal frames.

What it does well

Timed shots in one prompt. The prompt accepts bracketed timecodes such as [0-3s] and [3-6s], so you can script what happens in each stretch of the clip.
Sound design as part of the prompt. Describe dialogue, ambience and effects alongside the picture and they are generated together with it.
Reference images bound by name. Attach images and cite them in the prompt as <IMAGE_REF_0>, <IMAGE_REF_1> and so on, so each one is tied to the subject you meant.
A draft tier. 360p is a cheap, fast preview for checking a prompt before you pay for 1080p or 4K.

Prompting tips

Lead with the subject and the camera move, then the setting and light, then the sound. For a multi-beat shot, use timecodes rather than a long run-on sentence. When you use reference images, keep the prompt about what happens in the scene and let the references carry the look. Reference images cannot be combined with a first or last frame, so pick one approach per run.

Limits to know

Aspect ratio is 16:9 or 9:16. Duration is billed per second of output, so length and resolution are the two settings that move the price most. Because the model generates audio itself, anything you want to hear must be in the prompt.

Parameters

What Gemini Omni 1.1 accepts, in the order the playground shows it. The full reference, with an example for each language, is on the API page.

Parameter	Type	Default	Allowed	What it does
`promptrequired`	string	Cinematic drone shot gliding through mist…	—	Describe the scene, subjects, camera moves and sound design
`first_frame`	string	—	one image url	The video starts exactly from this image
`last_frame`	string	—	one image url	Optional ending frame
`reference_images`	string[]	—	up to 10 image urls	1-10 reference images
`duration`	integer	5	3–10	Video length in seconds (3-10)
`resolution`	enum	720p	360p · 720p · 1080p · 4k	Output resolution
`aspect_ratio`	enum	16:9	16:9 · 9:16	16:9 landscape or 9:16 vertical

Gemini Omni 1.1 pricing

Pay per second, priced by the settings you pick. These are our live rates; the strike-through is the published rate where one exists.

Setting	You pay	Published rate	Billed
360p	$0.042	—	per second of output
720p	$0.127	—	per second of output
1080p	$0.19	—	per second of output
4k	$0.38	—	per second of output

On top: $0.01 per request — Per-request model overhead; $0.002 per reference image — Per input image.

The exact price for your settings is on the Run button before you spend anything, and a failed run costs $0.00. Compare every model and plan.

Call Gemini Omni 1.1 from code

One POST returns a request id at once; poll it or take a webhook. Python, JavaScript, webhooks and every parameter are on the Gemini Omni 1.1 API page.

curlPOST /v1/models/gemini-omni-1.1

curl -X POST https://api.ezflows.io/v1/models/gemini-omni-1.1 \
  -H "Authorization: Bearer $EZFLOWS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Cinematic drone shot gliding through misty pine mountains at sunrise, god rays breaking through the fog, birdsong and gentle wind in the audio, photorealistic, highly detailed.",
    "duration": 5,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'

# 202 Accepted, at once — it never waits for the model:
# {"request_id": "req_8f2a41c9b70e4d15", "status": "queued",
#  "poll_url": "https://api.ezflows.io/v1/requests/req_8f2a41c9b70e4d15"}

Gemini Omni 1.1: questions

When should I use a first frame instead of reference images?

Use a first frame when the clip must begin exactly on a specific picture, such as a product shot or a storyboard frame. Use reference images when you want the model to carry a character, object or style into a scene it composes itself. The two cannot be used together.

Is the audio a separate step?

No. The soundtrack is generated with the video from the same prompt, so describe the sounds you want in the prompt itself.

How much does Gemini Omni 1.1 cost?

Gemini Omni 1.1 is priced by setting: $0.042 to $0.38 per second of output, depending on the options you choose. The exact price for your settings is shown on the Run button before you spend anything, and a failed run costs $0.00.

How long does a Gemini Omni 1.1 run take?

The measured average is ~53s, end to end. Longer outputs and higher settings take longer. The API is asynchronous: you get a request id at once, then poll it or receive a webhook when the result is ready.

How do I call Gemini Omni 1.1 from code?

Send a POST to https://api.ezflows.io/v1/models/gemini-omni-1.1 with your API key and the parameters as a JSON body. It answers 202 with a request id and a poll URL; poll it until the status is final, or register a webhook. Examples in curl, Python and JavaScript are on the API page.

Where are Gemini Omni 1.1 results stored, and for how long?

Results are saved to media.ezflows.io and kept for 90 days. Download anything you want to keep longer.