What to know:

  • Google unveiled Veo 3.1, an AI video generation model with richer audio, enhanced realism, and stronger narrative control.
  • Veo 3.1 adds audio to Flow’s key features like Ingredients to Video, Frames to Video, and Extend for seamless, lively clips.
  • The update enables tweaking shadows, lighting, and soon object removal, enhancing cinematic quality and natural-looking scenes.
  • Available in the Gemini app, Gemini API, and Vertex AI, it supports 1080p videos in landscape and portrait formats.

Google has launched Veo 3.1, the latest iteration of its AI-powered video model, bringing significant improvements to video generation and editing. Building on May’s Veo 3 release, the update offers more realistic clips, better adherence to user prompts, and enhanced audio output across generated videos.

Veo 3.1 introduces granular editing controls, letting users add objects to a video while seamlessly blending them into the clip’s existing style. Google also plans to enable object removal in Flow, giving creators more flexibility in post-production.

The model enhances existing Veo 3 features, including:

  • Adding reference images to guide character generation.
  • Using the first and last frame to produce AI-generated clips.
  • Extending videos based on the final frames of an existing clip.

With Veo 3.1, all these features now include audio, making the resulting clips more lively and immersive.

Google is rolling out Veo 3.1 across its Flow video editor, the Gemini App, and via Vertex and Gemini APIs for developers. Since Flow’s launch in May, users have created over 275 million videos on the platform, demonstrating the growing adoption of AI-assisted video tools.

This update positions Google’s video AI as a powerful tool for content creators, marketers, and developers looking to generate realistic, customizable video content quickly.