API Market / AI News / Gemini Omni 1.2 Flash: 15-second clips, edits in conversation, $0.07 per second

Gemini Omni 1.2 Flash: 15-second clips, edits in conversation, $0.07 per second

Published: November 6, 2026 · Official

Google DeepMind has released Gemini Omni 1.2 Flash, a video generation and editing model. It produces clips up to 15 seconds instead of 10, supports multi-turn conversational editing, and costs $0.07 per second of 720p video — 30% less than Omni 1.1 Flash. It is generally available on the Gemini API and Google Cloud.

What is new in Gemini Omni 1.2 Flash

Google DeepMind announced Gemini Omni 1.2 Flash on November 5, 2026. The model replaces Gemini Omni 1.1 Flash and brings several notable changes: longer clips, better reference adherence, multi-turn conversational editing, and more natural speech. According to Google, the price is cut by 30% to $0.07 per second of 720p video. A full 15-second 720p clip works out to about $1.05.

The model is natively multimodal: it takes text, images, short video clips, and audio as input and outputs video with audio. Google describes the Omni line as natively multimodal, with editing handled through the Interactions API. According to MarkTechPost, the previous clip does not need to be re-uploaded — it is passed through a previous interaction ID.

Editing in conversation and working with references

One of the main upgrades is multi-turn conversational editing. You can swap a product, background, lighting, or aspect ratio step by step instead of starting over. Google gives an example: a single generated clip changes its background turn by turn — from a striped pavilion tent to an oak forest, a stable with a white horse, a market fair, and a stone castle — while both characters, armor reflections, and camera framing stay intact.

Reference adherence is improved as well. The model combines character, product, or style images, short motion clips up to three seconds, and audio cues, holding those details from the first frame to the last. In a Google demo, one model photo plus front-and-back shots of three outfits became a 15-second virtual try-on sequence with 360-degree turns, preserving the model's identity and each garment's cut. According to Google, audio references on Google Cloud remain allowlist-gated.

Quality, speech, and camera control

Google notes improvements in speech generation: dialogue sounds more natural, lip-sync is clearer, and distinct multi-speaker conditioning is supported. In one example, the model received a conservatory scene image, four character, mouth, and one-second voice references, and routed each voice to the right character with expressive lip-sync and comedic timing across a 15-second scene.

Camera control is more predictable: complex continuous movements such as FPV-style flythroughs, dramatic zoom-outs, and orbits show fewer unwanted jump cuts and steadier scale. Google has not published benchmark scores for version 1.2. According to MarkTechPost, the company's claims rest on side-by-side demos against Omni 1.1 and on earlier results for the Omni Flash line, including 1,003 MovieGenBench prompts. Those results predate 1.2 and do not describe it.

Availability and partner feedback

Gemini Omni 1.2 Flash is generally available on the Gemini API and Google Cloud, and developers can work with it in Google AI Studio. In the Gemini app, 15-second video creation requires a Google AI Ultra subscription, while free users get some of the new capabilities in Flow and Google Vids. According to Android Authority, Google Vids offers up to three free 15-second videos per month, including reference images and improved speech narration. Google has not said when the model might reach AI Plus and Pro tiers.

Partners are already commenting on the model. invideo said it sticks to the prompt and that physics and sound feel real, and that 15 seconds means a full scene in one go. Photoroom uses Omni Flash to power its Video Ad Maker and Product Animator, and QuickFrame AI says 15 seconds at 1080p lets users build a complete connected-TV ad in a single generation. Google Cloud highlights enterprise-grade infrastructure, Provisioned Throughput, and an imperceptible SynthID watermark on outputs.

What to use right now

Google's new model can be assessed through its documentation and demos, while similar tasks — video generation and multimodal requests — are already covered by models in the API Market catalog. For video generation there is Seedance 2.0 Mini with per-second pricing, which fits quick script checks and short clips.

For the text side — scripts, prompts, shot descriptions — Gemini 3.8 Flash as a very fast Google model and Gemini 4 Argon with senior quality and long context are good options. One API Market key covers multiple models, with payment in rubles and no VPN. Keep an eye on the API Market model catalog as it updates with new versions.

Models you can use right now

ModelAPI Market price
Seedance 2.0 Mini
seedance-2.0-mini		$0.060 per request
Gemini 3.8 Flash
gemini-3.8-flash		input $0.07, output $0.35 per 1M tokens

Try these models on API Market

One API key for GPT, Claude, Gemini, DeepSeek, Grok and more. OpenAI-compatible endpoint, pay as you go, top-ups from about $0.15.

Get an API keyAll models

Connect in a minute

from openai import OpenAI

client = OpenAI(
    base_url="https://api-market1.shop/v1",
    api_key="sk-your-key",
)

reply = client.chat.completions.create(
    model="seedance-2.0-mini",
    messages=[{"role": "user", "content": "Hi! Tell me about yourself."}],
)
print(reply.choices[0].message.content)

FAQ

How much does Gemini Omni 1.2 Flash cost?

Google lists $0.07 per second of 720p video. That is 30% less than Gemini Omni 1.1 Flash. A full 15-second 720p clip works out to about $1.05 before additional charges.

What is the maximum clip length?

Up to 15 seconds natively, compared with 10 seconds for Omni 1.1 Flash. Duration can be set in one-second steps from 3 to 15 seconds.

What is conversational editing?

It means refining a clip step by step in a conversation: you can swap a product, background, lighting, or aspect ratio without regenerating the video from scratch. According to MarkTechPost, the previous clip is passed by ID rather than re-uploaded.

Where is the model available?

On the Gemini API and Google Cloud. In the Gemini app, 15-second video creation is available to Google AI Ultra subscribers, while some capabilities are free in Flow and Google Vids.

Are there published benchmarks for the model?

No. According to MarkTechPost, Google has not published benchmark scores for version 1.2; its claims rest on side-by-side demos against Omni 1.1 and on results for earlier versions of the line.

Sources

More news

Русская версия