Google's video generation AI 'Veo 3.1' has been updated to improve character consistency and support 4K upscaling

Google announced an update for its video generation AI, Veo 3.1 , on January 13, 2026. The update includes improvements to the ability to create videos based on reference images, improved character consistency, and the addition of 4K upscaling.

Veo 3.1 Ingredients to Video: New video generation model updates
https://blog.google/innovation-and-ai/technology/ai/veo-3-1-ingredients-to-video/

Veo 3.1 includes a new feature called 'Input an image and move an object in the image.' This feature has been improved to generate dynamic and expressive videos with just a few prompts.

Veo 3.1 Updates - Bring more creativity and expressiveness into your videos - YouTube


Character consistency has also been improved, so that the face and body shape of a person can be maintained even when performing operations such as 'inputting a person, clothing, and background and generating a video.'

Veo 3.1 Updates - Maintain identity consistency for your characters - YouTube


The consistency of the backgrounds and placed objects has also been greatly improved.

Veo 3.1 Updates - Achieve background and object consistency - YouTube


Furthermore, it also supports 4K upscaling, allowing you to generate high-definition videos using Veo. It also supports vertical video output.

Veo 3.1 Updates - Generate videos in 1080p and 4K with state-of-the-art upscaling - YouTube


The updated Veo 3.1 is available on YouTube Create , the Gemini app, Flow, Gemini API, and Google Vids.