Gemini Robotics 2.5 Targets More Dexterous and Reliable Robots
Google DeepMind has announced Gemini Robotics 2.5, a vision-language-action model for physical intelligence. According to DeepMind, the model is intended to enhance robots’ dexterity, reliability, and safety when operating in the real world.
The release extends the use of multimodal AI systems beyond information processing and into direct physical interaction, where models must connect visual understanding with robotic actions.