Google DeepMind's Gemini Robotics ER 2 Enables Full Body Control

Google DeepMind has released Gemini Robotics ER 2, an embodied reasoning (ER) model available on Google AI Studio and Gemini Enterprise Agent Platform. This vision-language model (VLM) allows robots to communicate, understand the physical world, and plan multi-step tasks. DeepMind also introduced Gemini Robotics On-Device 2, an optimized VLA that runs locally on robots and can adapt to new robotic embodiments with just a few hours of data, including controlling complex five-fingered hands for delicate actions.