Gemini Robotics On-Device brings AI to local robotic devices

June 24, 2025 | Source: Google DeepMind Blog

Tags: Gemini Robotics, DeepMind, on-device AI, robotics, VLA model, task generalization, SDK

Gemini Robotics On-Device introduces an efficient AI model for local robotic devices, enhancing dexterity and task adaptation without reliance on external data networks.

Details

Gemini Robotics On-Device is a new robotics model from DeepMind that operates locally on robotic devices, offering general-purpose dexterity and rapid task adaptation. This model, part of the Gemini 2.0 suite, is designed for bi-arm robots and optimized for low-latency performance, making it suitable for environments with limited connectivity. The accompanying Gemini Robotics SDK allows developers to evaluate and adapt the model for specific tasks with minimal demonstrations. The On-Device model demonstrates strong generalization capabilities, effectively following natural language instructions and performing complex tasks such as unzipping bags and folding clothes.