Google has released Gemini Omni 1.2 Flash, an updated AI video model that makes longer clips, keeps characters and products more consistent, and costs 30% less than the version before it. The model is now generally available through the Gemini API, Google Cloud and Gemini for business. If you make short videos for ads, social posts or product demos, this update is aimed squarely at you.
What Is New in Gemini Omni 1.2 Flash
The biggest change is length. The new model can create and edit videos natively up to 15 seconds long, up from 10 seconds with Gemini Omni 1.1. You can also pick an exact length in one-second steps, anywhere from 3 to 15 seconds.
Google says that extra time is enough for a full social ad, a short product explainer or a complete little story in one generation. That means fewer clips stitched together and fewer awkward cuts.
Better consistency from start to finish
Google says the model holds onto reference details much better than before. You can feed it character, product or style images, short motion clips of up to three seconds, and audio cues. Faces, clothing, packaging text and brand colors are meant to stay the same across camera angles, with less of the drifting and morphing that often ruins AI video.
Gemini Omni 1.2 Flash Pricing and Availability
The price drop is a big part of this release. Google cut the cost by 30% compared with Omni 1.1, bringing it to $0.07 per second for 720p video. A full 15-second 720p clip would therefore cost about $1.05 before any upscaling.
| Feature | Gemini Omni 1.1 | Gemini Omni 1.2 Flash |
|---|---|---|
| Max native clip length | 10 seconds | 15 seconds |
| Price change | Baseline | 30% lower |
| 720p price | About $0.10 per second | $0.07 per second |
| Multi-turn editing | Not listed | Yes, across full clips |
| Availability | Earlier release | Generally available |
Google also suggests a cheaper way to work. You can draft fast at 360p or 720p, fix details with follow-up edits instead of starting over, and then upscale your best clips to 1080p or 4K.
Where you can use it
This is the first time Gemini Omni is generally available, backed by Google Cloud’s enterprise infrastructure. Developers can test it in Google AI Studio and the Gemini API, while businesses can use it through the Google Cloud console. Google says it also shows up in Google Vids, Flow and Gemini for business, and that every output carries an invisible SynthID watermark.
Editing Videos by Chatting
Another key feature is multi-turn conversational editing. Instead of writing a brand new prompt every time, you can refine the same clip step by step. Google’s examples include swapping a product, changing the background, adjusting the lighting or weather, and switching the aspect ratio.
This works across the full 15-second clip, which makes it easier to create several ad versions for different audiences. Here are the main editing and quality upgrades Google lists:
- Swap products, backgrounds, lighting or weather while keeping the rest of the scene intact
- More natural dialogue, clearer lip-sync and distinct voices for multiple speakers
- Smoother camera moves like drone flythroughs, orbits and zoom-outs with fewer jump cuts
- Better first-to-last frame consistency when linking two keyframes or extending a clip
Google is honest that the model still has limits. It says work continues on fast sports action, rapid script changes and complex physics, which remain hard for AI video tools in general.
What This Means for You
For marketers, small businesses and creators, the mix of longer clips and lower prices is the real story. Here is how to think about it:
- Test cheap drafts first. Start at 360p or 720p, fine-tune with edits, and only upscale the winners.
- Use your own reference images. Product photos and character sheets can now guide whole clips, which helps with brand consistency.
- Plan for full-length ads. Fifteen seconds fits many social ad formats in a single generation.
- Remember the watermark. SynthID marks every output as AI-made, which matters for disclosure and trust.
If you are comparing options, see how other models stack up in our video quality comparison and our look at the Kling 4 model. Our guide to video marketing tools can help you build a workflow around these clips.
Struggling with odd results? Our explainer on distorted AI faces covers common fixes. For more on Google’s latest models, read about the Gemini 4 Argon launch or browse our AI news hub.
Sources
Primary source: Google DeepMind’s Gemini Omni 1.2 Flash announcement.
Conclusion
Gemini Omni 1.2 Flash makes Google’s video model longer, steadier and cheaper all at once. The jump to 15-second clips, chat-style editing and a 30% price cut make it a practical choice for anyone producing video at scale. It still struggles with fast action and complex physics, so it will not replace a film crew. For quick ads, explainers and social clips, though, it looks like a practical upgrade worth testing.
