Google DeepMind Launches Gemini Robotics ER 2, an Embodied Reasoning Model for Robots
Google DeepMind has released Gemini Robotics ER 2, their most capable 'embodied reasoning' model for robotics. This vision language model (VLM) acts as a high-level brain, allowing robots to communicate with humans, understand the physical world, and plan multi-step tasks. It can coordinate with lower-level vision-language-action (VLA) models for motor execution and natively call tools like Google Search. Gemini Robotics ER 2 significantly improves safety by enhancing human proximity detection and enabling multi-robot collaboration.
This model drastically improves robot autonomy and safety, enabling more complex, multi-step tasks and safer human-robot interaction in shared environments. Its ability to plan and adapt in real-time is a significant leap for robotic intelligence.
Learn one new AI thing every day.
Daily Deck sends you seven plain-English cards like this every morning. Free.
Start free