Gemini 3: Interact with a virtual OS by simply drawing

Gemini 3’s improved spatial reasoning also powers intelligent screen understanding of desktop, mobile and OS screens delivering significant performance improvement for computer use agents. In this app built in Google AI Studio, see how the understands the intent of user actions based on screen annotations to take actions instructed by the user.

Find out more about Gemini 3: https://deepmind.google/models/gemini/
___

Subscribe to our channel https://www.youtube.com/@googledeepmind
Find us on X https://twitter.com/GoogleDeepMind
Follow us on Instagram https://instagram.com/googledeepmind
Add us on Linkedin https://www.linkedin.com/company/deepmind/