Veo

Google DeepMind's dedicated video model, now at Veo 3.1. Describe a scene or hand it reference images and it produces an 8-second clip with native sound - dialogue, effects and music generated together with the picture - at 1080p or 4K. Physics, lighting and camera moves look convincingly real, and characters stay consistent from shot to shot. Where to use it: Flow (Google's filmmaking tool), Google AI Studio, Google Vids and the Gemini API for developers. In the Gemini app itself, video is now mostly handled by Gemini Omni, a newer model that lets you fix a clip in conversation instead of regenerating it. Serious use isn't cheap, but if you need AI video that doesn't look like AI video, Veo is still one of the strongest options.

Something here out of date? Tell me