Google has introduced Veo 3.1, the latest version of its AI-powered video generation model, bringing upgraded audio output, refined editing controls, and improved image-to-video generation. The company said Veo 3.1 builds on May’s Veo 3 release, producing more realistic clips that align more precisely with user prompts.
A key enhancement in this update is the ability to add an object into a video and have it seamlessly blend with the clip’s original style. Google noted that users will soon gain the option to remove existing objects from videos within its Flow editor as well.
Veo 3 already offered robust tools like reference image-driven character editing, AI-generated video clips based on specified first and last frames, and extensions of existing videos using their final few frames. With Veo 3.1, these capabilities are now paired with built-in audio generation, creating livelier, more immersive clips.
The new model is being rolled out across Google’s Flow video editor, the Gemini App, and through Vertex and Gemini APIs. According to Google, since Flow’s debut in May, users have already created over 275 million videos using the platform.
