Google has introduced Gemini Robotics 2, a new AI system that enables robots to handle more complex tasks, adapt to changing environments, and collaborate with other robots. The release includes three sub-models, one of which is now publicly available to developers through the Gemini Live API.
The lead model, Gemini Robotics ER 2, is an upgraded “embodied reasoning” system that DeepMind says is a major improvement over the previous 1.6 version. With this update, Google aims to move beyond narrowly programmed machines toward a general-purpose robot—what its scientists call “physical AGI”—with better control of humanoid bodies and hands.
Comments