Google DeepMind Unveils Gemini Robotics 2 for Full-Body Humanoid Control

Google DeepMind has launched Gemini Robotics 2, a suite of AI models enabling full-body coordination, multi-step reasoning, and multi-robot collaboration for humanoid robots. The flagship Gemini Robotics 2 is a vision-language-action model that translates visual and language inputs into motor commands across an entire humanoid body, demonstrated on Apptronik's Apollo 2 robot. This advancement allows robots to perform complex tasks like walking, retrieving objects, and precise placement, adapting in real-time to environmental changes.