📰 Key Takeaways

Google has launched Gemini Omni 1.2 Flash, now generally available in the Gemini API and Gemini for business. It’s an upgraded video generation model that focuses on speed and low cost, and adds native 15-second video generation, multi-turn conversational editing, and stronger adherence to reference materials. Pricing is 30% lower than Omni 1.1, at $0.07 per second for 720p video.

On cost, the officially recommended workflow is to produce a quick draft at 360p or 720p first. If you only need to fix one detail afterward, use a follow-up editing instruction instead of regenerating the whole video, and then upgrade the version you like to 1080p or 4K.

Video length goes from 10 seconds in Omni 1.1 to 15 seconds, and you can set the duration anywhere between 3 and 15 seconds in one-second increments. That suits single-pass needs like product explainers or complete social ad clips.

For reference materials, you can guide generation by combining character, product, or style images, motion clips of up to three seconds, and audio prompts. The model keeps faces, clothing, packaging typography, and brand colors consistent across different camera angles. In the official demo, a single model photo plus front and back photos of three outfits produced a 15-second virtual try-on video, complete with a 360-degree spin and match cuts, while preserving the model’s identity and each garment’s cut, drape, and back details.

Consistency has improved too. There’s less drift and distortion of characters, objects, and scenes from the first frame to the last, and shots stay consistent whether you’re interpolating between first and last keyframes or extending a clip. The official post also uses a five-layer animation prompt for an exhibition poster to show that 1.2 follows the reference image more closely than 1.1.

For editing, it supports multi-turn conversation. You can swap the product or background, change the lighting or weather, or switch the aspect ratio turn by turn without starting over, and it works on full 15-second clips. The second half of the original article was truncated, so please see the source link for the full details.

💬 JudyAI Lab Take

Google’s Gemini Omni 1.2 Flash is now generally available in the Gemini API. It natively supports 15-second video generation, and the price is another 30% lower than the previous generation. Video generation is moving from “flashy demo” to a stage where you can actually calculate costs and fit it into a workflow, and that’s worth paying attention to.

This case reflects two design ideas. The first is building cost control into the workflow: the official advice is to make a quick draft at 360p or 720p, fix local problems with follow-up editing instructions instead of redoing the whole clip, and only upgrade to 1080p or 4K once you’re happy. At $0.07 per second for 720p, this “try cheap first, polish later” order is what really saves money. The second is that consistency and multi-turn editing have become the core selling points. The model keeps faces, clothing, packaging fonts, and brand colors consistent across camera angles, and it can swap products, backgrounds, lighting, or aspect ratios turn by turn. For AI builders, the focus of product design is shifting from “it has to be perfect in one shot” to an interaction model where you can keep revising through conversation, and the user experience will change along with it.

If your product or workflow uses video generation, try the official order once: draft at low resolution, fix details with editing instructions, upgrade the quality last, and record the actual cost of each step.

📅 Source Info

🔗 Further Reading