Gemini Robotics 2.5 accelerates real-world general dexterity
Google DeepMind
Google DeepMind has introduced Gemini Robotics 2.5, a vision-language-action model designed to enhance real-world robot dexterity and safety.
Google DeepMind has announced Gemini Robotics 2.5, its most capable vision-language-action model to date. The system is engineered to significantly improve robot dexterity, reliability, and safety in real-world applications. This advancement builds on multimodal AI capabilities to better integrate robotic perception and physical action.
Why it matters: The development represents an advancement in vision-language-action models that could improve the practical dexterity, safety, and reliability of robots in real-world settings.






