<ul><li>Google DeepMind is launching two new AI models to improve robots' performance in physical tasks.</li><li>Gemini Robotics is a vision-language-action model that can understand new situations without prior training.</li><li>Advancements in generality, interactivity, and dexterity make Gemini Robotics more capable, responsive, and adaptable.</li><li>Google DeepMind is also introducing Gemini Robotics-ER, an advanced visual language model for embodied reasoning and complex world understanding.</li></ul>

Google DeepMind’s new AI models help robots perform physical tasks, even without training

Discover more