Google unveils Gemini Robotics-ER 1.6, a model designed for robotic applications.

CoinFeed reported on April 14th that, according to the Google DeepMind website, Google has released Gemini Robotics-ER 1.6, an embodied reasoning model for robotics. Compared to Gemini Robotics-ER 1.5 and Gemini 3.0 Flash, it significantly improves spatial and physical reasoning tasks such as pointing and localization, object counting, and multi-view success detection, and adds the ability to read industrial instruments. This model can serve as a high-level decision-making hub for robots, natively calling tools such as Google Search and VLA. Through agentic vision, it first magnifies key areas and then combines pointing and code calculations to achieve high-precision instrument reading.