Data as of Sep 29, 2026 · Based on 16,881 AI responses · See how Parse measures this
Gemini Robotics On-Device 2 is a state-of-the-art on-device Vision-Language-Action model for general-purpose robotic manipulation that runs efficiently on local devices. It accepts text instructions, images from the robot's environment, and proprioceptive data as inputs, and outputs robot actions as numerical values. Built on the on-device Gemma models, it supports diverse tasks and robot types and is currently available to a select group of trusted testers.
<1%No change
of AI answers about Gemini Robotics On-Device 2 and its rivals. Last 30 days
“Gemini Robotics On-Device 2 is the interesting piece if latency, connectivity, or privacy requires inference on the robot rather than in the cloud.”
Last 30 days
AI mentioned Gemini Robotics On-Device 2 in <1% of answers about Gemini Robotics On-Device 2 and its rivals in the last 30 days.
ChatGPT Search mentions Gemini Robotics On-Device 2 in <1% of its answers to Gemini Robotics On-Device 2's market questions, Google AI Mode in 0%.
