APIMODELSAPIMODELS

Gemini Omni API pricing: every Omni video model compared

Updated 2026-11-21

TL;DR On apimodels.app the newest Gemini Omni, Omni 1.2 Flash, costs $0.045 per second at 720p and $0.052 at 1080p, against Google’s listed ~$0.07 and ~$0.104, so an 8-second 720p clip is $0.36. Omni 1.1 Flash adds 4K, video editing and up to 7 reference images from $0.07 per second, and the two per-clip Omni Flash tiers start at $0.1125 (Lite) and $0.35 (Stable). Every clip carries synced audio, failed generations are not charged, and one API key calls all of them.

Model page: Gemini Omni 1.2 Flash API — pricing, specs and the playground

Google sells Gemini Omni through the Gemini API and Google Cloud and bills it by output tokens, which its pricing page converts to about $0.07 per second at 720p for Omni 1.2 Flash. From 3 December 2026 Google serves requests for Omni 1.1 Flash with 1.2 at the 1.2 price. On APIMODELS each Omni model is its own model id on one video endpoint, billed per second or per clip as listed below, with no Gemini API key or Google Cloud project to set up.

Gemini Omni price by model

USD on apimodels.app. Per-second models bill each second of output; per-clip models charge one price per clip by length. All of them generate synced audio.

Model idBilling720p1080p4KLengthsPick it for
gemini-omni-1.2-flashper second$0.045$0.052—3–15 sNewest; longest clips; up to 3 reference images or first + last frame
gemini-omni-1.1-flashper second$0.07$0.10$0.204 / 6 / 8 / 10 s4K, up to 7 reference images, saved voices and characters, video editing at $0.70 flat
gemini-omni-videoper clip$0.35–$0.60same as 720p$0.70–$1.004 / 6 / 8 / 10 sOmni Flash (Stable): one flat price per clip, video-to-video, voices and characters
gemini-omni-video-liteper clip$0.1125–$0.225same as 720p—4 / 6 / 8 / 10 sOmni Flash (Lite): lowest per clip; leaves a faint patch in the bottom-right corner
gemini-omni-flashper second$0.06$0.065$0.126 / 8 / 10 sOmni Flash (Gemini): text, 1 or 3 reference images, or a video to edit

Omni 1.2 Flash against Google’s price

Per second of output. Google’s figures are the per-second equivalents on its AI Studio pricing page, checked on 21 November 2026.

ResolutionGoogle (Gemini API)APIMODELSDifference8 s / 15 s on APIMODELS
720p~$0.07$0.045−36%$0.36 / $0.675
1080p~$0.104$0.052−50%$0.416 / $0.78

What Omni 1.2 Flash does not do here

Google’s Omni 1.2 Flash also offers 360p, 4K, video input and extension. On APIMODELS, gemini-omni-1.2-flash takes text, up to 3 reference images, or a first frame with an optional last frame (3 to 10 seconds when both are set), at 720p or 1080p; a first frame cannot be combined with reference images. For 4K or editing an existing video, use gemini-omni-1.1-flash or gemini-omni-video.

No Omni model succeeds on every request. Failed generations are not charged.

What is charged and what is not

There is no free tier and no monthly fee. Usage comes out of a prepaid balance; top-ups start at $10 and can be paid by card, PayPal, Alipay or USDT. New accounts registered with a personal email address or Alipay get a one-time $0.10 trial credit, which is less than the shortest Omni clip (3 seconds of Omni 1.2 Flash at 720p is $0.135).

When this is not what you need

If you want to make Omni videos in the Gemini app rather than call the model from code, use Google’s app; this page is about the API.

If you need 360p drafts, or clips extended past 15 seconds, Google’s own API offers them and these endpoints do not.

Início rápido

cURL

curl -X POST https://api.apimodels.app/v1/video/generations \
  -H "Authorization: Bearer $APIMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-omni-1.2-flash",
    "prompt": "A street musician plays violin at dusk, the crowd slowly gathers",
    "resolution": "720p",
    "duration": 8
  }'

Python

import requests
H = {"Authorization": "Bearer YOUR_APIMODELS_KEY"}
r = requests.post("https://api.apimodels.app/v1/video/generations", headers=H, json={
    "model": "gemini-omni-1.2-flash",
    "prompt": "The sneaker from the reference spins on a pedestal, studio light",
    "images": ["https://example.com/sneaker.jpg"],       # up to 3 reference images
    "resolution": "1080p",
    "duration": 6,
})
print(r.json())  # poll GET /v1/video/generations?task_id=...

Perguntas frequentes

How much is Gemini Omni Flash?

It depends on the version. Omni 1.2 Flash is $0.045 per second at 720p and $0.052 at 1080p; Omni 1.1 Flash is $0.07, $0.10 and $0.20 per second at 720p, 1080p and 4K; the per-clip Omni Flash tiers run from $0.1125 (Lite, 4 seconds) to $1.00 (Stable, 4K, 10 seconds).

What is the Gemini Omni price per second?

Google lists Omni 1.2 Flash at about $0.07 per second at 720p and $0.104 at 1080p on the Gemini API, converted from its token price. On apimodels.app the same model is $0.045 and $0.052 per second, so a 15-second 720p clip is $0.675.

Is Gemini Omni free?

Not through the API. There is no free tier here: each clip is billed from a prepaid balance (top-ups from $10), and failed generations cost nothing. New accounts registered with a personal email address or Alipay get a one-time $0.10 trial credit, which is less than the shortest Omni clip.

Do I need a Gemini API key to use Gemini Omni?

Not on apimodels.app. You call every Omni model with one apimodels.app key on POST /v1/video/generations; there is no Gemini API key, Google Cloud project or billing account to set up. The same key also calls Veo 3.1, Seedance, Kling and the other video, image and language models on the site.

Relacionado

Presente em
APIMODELS on Product HuntAPIMODELS featured on ToolifyAPIMODELS featured on There's An AI For That