Google Omni 1.2 Flash

Gemini Omni 1.2 Flash is Google DeepMind's video generation and editing model, now generally available. Key features: • Clips up to 15 seconds, up from 10 seconds in the previous version • Multi-turn editing that refines a clip without regenerating it from scratch • Steadier characters and consistent visual styles across shots • Clearer lip-sync and smoother camera movement • Draft at 360p or 720p, then upscale to 1080p or 4K • Available in the Gemini API, Google Cloud, Flow, and Vids Omni 1.2 Flash is built for speed and cost efficiency. Google prices 720p output at $0.07 per second, about 30% less than Omni 1.1. Because you can draft at lower resolutions and upscale only the final version, you can iterate on ideas quickly and keep production costs under control. Multi-turn editing lets you adjust lighting, motion, or dialogue in follow-up instructions without losing the parts of the clip that already work. The improvements to consistency and motion address common frustrations with generated video. Characters and styles hold together more reliably between scenes, lip movement lines up more clearly with speech, and camera pans and moves look smoother. Together, these changes make generated footage more usable in real projects rather than only as rough experiments. This model suits developers building video features into apps, marketing teams producing short social clips, creative studios prototyping storyboards, and educators or product teams who need quick explainer videos. Teams already working in Google Cloud, Flow, or Vids can start using it directly within their existing tools.