Google Deepmind unveils Gemini Robotics 2 to power robots of all shapes from tabletop arms to humanoids

Google Deepmind has introduced Gemini Robotics 2, an updated vision-language-action model intended for embodied AI applications. The system is designed to generalize across various robotic form factors, ranging from small tabletop manipulators to full-body humanoid machines.

The release highlights a new reasoning layer aimed at improving task planning and execution within physical environments. This advancement suggests a move toward more versatile AI agents capable of handling complex, real-world interactions without extensive retraining for specific hardware.

By unifying control across different robot shapes, Deepmind aims to lower the barrier for deploying AI in physical settings. This development aligns with broader industry trends focusing on multimodal models that bridge digital intelligence with physical action.

This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.