google-deepmind

In a new interview, Google DeepMind's Carolina Parada discusses the 'agentic' two-part brain of Gemini Robotics 1.5, its 'surprising' ability to transfer skills between different robots, and why she believes the home will be 'one of the last frontiers' for the technology.

Google DeepMind has unveiled Gemini Robotics 1.5, a new AI framework designed to move robots beyond simple command-following to become 'physical agents' that can reason, plan, and learn skills across different hardware, like Apptronik's Apollo humanoid.

Google DeepMind has launched Gemini Robotics On-Device, an efficient vision-language-action model designed to run locally on robotic hardware. This development aims to reduce latency, improve robustness in varied environments, and enable advanced dexterous manipulation for a new generation of autonomous robots capable of fast task adaptation.