Google DeepMind Unveils Gemini Robotics 2

Google DeepMind has released Gemini Robotics 2, a suite of three AI models designed for whole-body robot control, embodied reasoning, and on-device execution. The platform enables humanoid and dual-arm robots to perform complex, multi-step tasks, including fine finger manipulation and multi-robot collaboration.

The models are currently available to developers in early-access form via the Gemini API, Google AI Studio, and private preview on the Gemini Enterprise Agent Platform.