Google unveils Veo 3.1 — a more realistic and cinematic AI video generator with sound

Google has introduced Veo 3.1, the latest version of its AI-powered video generation model featuring major upgrades in realism, adherence to text prompts, and sound design.

Compared to the version released in May, Veo 3.1 produces more lifelike visuals and can now add audio to generated scenes.

Users can not only create clips from text descriptions but also extend existing videos beyond their final frame or link characters to reference images for greater consistency.

According to Google, the new algorithms make videos significantly more vivid and cinematic. The model is already integrated into the Flow video editor and is also available through the Gemini app, Gemini API, and the Vertex AI platform.

Flow users have additionally gained the ability to remove objects from videos using AI tools. Since Flow’s launch in May, users have generated more than 275 million videos, a number that Google says continues to rise rapidly.