Gemini Omni 1.2 Flash: 15-Second AI Video, Price, Features

Google DeepMind made Gemini Omni 1.2 Flash generally available on 5 November 2026. It is a video generation and editing model that now makes clips up to 15 seconds long, and Google says it costs 30% less than Gemini Omni 1.1.

Quick summary

• Clip length: 3 to 15 seconds, in one-second steps (up from 10 seconds).
• Price: $0.07 per second for 720p video.
• New: multi-turn conversational editing, better reference following, clearer lip-sync.
• Output: draft at 360p or 720p, upscale to 1080p or 4K.
• Marking: outputs carry an invisible SynthID watermark.

What is new in Gemini Omni 1.2 Flash?

Google highlights five changes. Clips are longer. The model holds onto faces, clothing, packaging text and brand colours better when you give it reference images, short motion clips of up to three seconds, or audio cues. You can edit a clip over several turns, such as swapping the background or lighting, instead of regenerating it. Speech sounds more natural with distinct multi-speaker voices. And camera moves like orbits and FPV flythroughs have fewer unwanted jump cuts. Google admits fast sports action, rapid script changes and complex physics still need work.

How much does it cost?

Google lists $0.07 per second for 720p. A 15-second 720p clip therefore works out to about $1.05 at that rate, simple arithmetic from the published figure. The cost-saving workflow Google suggests is to draft cheaply at low resolution, fix details with a follow-up edit, and only upscale the best result.

Where can you use it?

• Developers: the Gemini API and Google AI Studio.
• Businesses: Google Cloud, with Provisioned Throughput available.
• Everyone with Google apps: Flow for scenes and edits, Google Vids (15-second clips at no extra cost with a Google account or Workspace), Google Ads asset tools, and the Gemini app, which is rolling out to Google AI Ultra subscribers first.

Google did not state India-specific pricing or availability in this announcement, so check your own account for regional rollout.

Why does it matter for creators and game developers?

Consistent characters and products across a 15-second shot make short trailers, product explainers and social ads far more practical to prototype. It also fits the trend of AI tools moving into game and PC workflows, as we saw with Unreal Engine's AI push. For the other big Google model, read our Gemini 4 Carbon guide.

Verdict

This is an incremental but practical upgrade: longer clips, cheaper output and editing without restarting. Independent side-by-side tests against rival video models are not yet available, so treat Google's own examples as marketing until reviewers publish comparisons.

How does 1.2 Flash differ from Omni 1.1?

Google's own comparison shows 1.2 matching reference images and first-to-last frame transitions more closely than 1.1, for example connecting an exterior start frame and an interior end frame in one continuous camera push.

Is the output labelled as AI?

Yes. Outputs carry an imperceptible SynthID watermark, Google's tag for AI-generated media, which helps platforms and tools detect synthetic video. It is invisible to viewers, so it does not replace clear disclosure when you publish.

Related articles

Gemini 4 Carbon: Access, Price and How It ComparesChatGPT Images 2.6 Adds GIF Creation and Two New API ModelsClaude Duet: Waitlist, India Access, Price and FeaturesChatGPT Search Is 20% of Google's Volume: What OpenAI Said