Gemini Omni API: Gemini Omni 1.2 Flash Video Pricing, Access and Calls, With or Without a Google Account (2026)

Gemini Omni 1.2 Flash (gemini-omni-1.2-flash) is Google's current video model in the Gemini API: text, images or video in, up to 15 seconds of video with native audio out, at about $0.07 a second at 720p on the paid tier, with no free tier. Gemini Omni 1.1 Flash, the previous generation, is billed at about $0.10 a second, and from December 3, 2026 every request to it is routed to 1.2 Flash at 1.2's rates.

This guide covers what the Omni models do, what Google charges by resolution, how to get access from Google, how to call Omni without a Google account, and which route fits whom.

A note on who wrote this: GenMagic is one of the routes compared here; its prices come from its live catalog, and on Gemini Omni they are above Google's own. Google's figures were read on November 10, 2026 on the Gemini API pricing page. For Veo 3.1, which left the Gemini API on October 22, see our Veo 3.1 API guide.

Table of contents

1. The short answer
2. The Omni models
3. What Google charges
4. Getting access from Google
5. Gemini Omni without a Google account
6. Which route to pick
7. Questions

The short answer

• Gemini Omni 1.2 Flash makes and edits video up to 15 seconds with native audio, from text, images or video. Google bills it per output token: 5,792 tokens a second at 720p, $12 per million video tokens, so about $0.07 a second at 720p.
• By resolution on Google: about $0.023 a second at 360p, $0.07 at 720p, $0.104 at 1080p and $0.209 at 4K (1080p and 4K are upsampled).
• No free tier: Omni runs only on the Gemini API's paid tier, which needs a billing account.
• Omni 1.1 Flash costs about $0.10 a second at 720p; from December 3, 2026 its requests route to 1.2 Flash.
• Without a Google account, GenMagic runs Gemini Omni 1.2 Flash at $0.10 a second at 720p with sound, above Google's price, from one balance with every other model.

The Omni models

Gemini Omni is not Veo: it is a Gemini model that generates video, called through the Gemini API and billed per output token. Google's AI plans now name Gemini Omni Flash as their video model, so Omni is also what many people get in Google's own apps.

What Google charges

On the Gemini API's paid tier, as Google's pricing page reads on November 10, 2026:

Input (your prompt, images or source video) costs $1.50 per million tokens. Gemini Omni 1.1 Flash is billed at $17.50 per million video tokens, about $0.10 a second at 720p, until its requests move to 1.2 Flash on December 3. On the paid tier, Google says your data is not used to improve its products.

Getting access from Google

1. A Google account and a Gemini API key from Google AI Studio.
2. The paid tier: Omni has no free tier, so the key's project needs billing turned on.
3. Code that sends the request and collects the video when it is ready.

That is the lowest price per second, and the right route if you already run on Google's cloud or need Omni's video editing at volume.

Gemini Omni without a Google account

On GenMagic, Gemini Omni 1.2 Flash sits behind the same two calls as every other video model: one starts a clip, the other asks whether it is ready. The balance pays once, when the finished clip comes back, and a render that fails is not charged.

curl https://genmagic.co/api/v1/videos \
  -H "Authorization: Bearer $GENMAGIC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-omni-1.2-flash",
    "prompt": "A lighthouse keeper climbs the spiral stairs at night with a lantern, wind howling outside, the beam sweeping across the sea. Slow tracking shot.",
    "duration": 10,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'

Then poll GET https://genmagic.co/api/v1/videos/<id> until the status is completed; the answer carries the hosted clip and its cost. Every field and model is in the developer docs.

What a clip costs on GenMagic today:

The same models run in the studio with no code, and from n8n, Zapier, Make, Claude and ChatGPT on the same balance. Every Gemini Omni model and price: Gemini Omni pay as you go.

Which route to pick

• You build on Google's cloud or need the lowest price per second: the Gemini API's paid tier.
• You want Omni without a Google billing setup, or next to Veo, Kling and Seedance on one key: GenMagic, at a higher price per second.
• You still call gemini-omni-1.1-flash: from December 3 Google routes it to 1.2 Flash; on GenMagic, choose 1.2 Flash for the longer clips (up to 15 seconds) and its lower price.

Questions

How much does the Gemini Omni API cost? On Google's paid tier, Gemini Omni 1.2 Flash is about $0.07 a second at 720p ($12 per million video tokens at 5,792 tokens a second), $0.104 at 1080p and $0.209 at 4K, plus $1.50 per million input tokens.

Is there a free tier for Gemini Omni? No. Google lists no free tier for Omni input or output; the paid tier needs billing.

What happens to Gemini Omni 1.1 Flash? From December 3, 2026, Google routes every request to gemini-omni-1.1-flash to gemini-omni-1.2-flash, billed at 1.2's rates.

Is Gemini Omni the same as Veo 3.1? No. Omni is a Gemini model that makes and edits video, billed per token; Veo 3.1 is Google's separate video model, now on its cloud platform. Both run on GenMagic: Veo 3.1 API guide.