Google DeepMind Launches Gemini Robotics 2 for Whole-Body AI Control
Google DeepMind has introduced Gemini Robotics 2, an advanced vision-language-action (VLA) model designed to provide whole-body intelligence for robots. This model converts visual and language inputs into motor control, allowing for the control of humanoids and bi-arm robots with enhanced dexterity. Gemini Robotics 2 enables robots to reason through movements, perform multi-robot collaboration, and adapt to new robot bodies within hours. The model can run locally on devices, emphasizing efficiency and adaptability.
Gemini Robotics 2 represents a major leap in robotics, offering highly adaptable and intelligent control over complex robotic systems, which could accelerate the development of versatile robots capable of performing intricate tasks in diverse environments.
Learn one new AI thing every day.
Daily Deck sends you seven plain-English cards like this every morning. Free.
Start free