Google has launched Gemini Robotics-ER 1.6, a major update to its reasoning-first artificial intelligence model designed to enhance embodied reasoning in robots. Launched today, this upgrade enables robots to better understand their physical environments through advanced spatial and multi-view capabilities. Unlike previous iterations that merely followed instructions, this model allows physical agents to reason about complex scenarios, bridging the gap between digital intelligence and real-world action. The new system specializes in critical robotics functions, including visual and spatial understanding, task planning, and success detection. It operates as a high-level reasoning brain for robots, capable of executing tasks by natively calling external tools such as Google Search, vision-language-action models, or user-defined third-party functions. This architecture allows robots to interpret their surroundings more deeply, such as navigating intricate facilities or identifying specific objects within a cluttered workspace. Compared to earlier versions, Gemini Robotics-ER 1.6 delivers significant improvements in spatial and physical reasoning. Key enhancements include more accurate pointing, counting, and success detection. A notable new capability is instrument reading, which empowers robots to interpret complex gauges and sight glasses. This feature was developed through close collaboration with Boston Dynamics, a leading robotics partner, to address real-world industrial needs where precise measurement interpretation is vital. The release marks a step toward greater autonomy for next-generation physical agents. By refining the model's ability to process visual data and plan tasks effectively, Google aims to make robots more useful in both daily life and industrial settings. The model serves as a robust foundation for developers building advanced robotic applications that require a deep understanding of the physical world. Gemini Robotics-ER 1.6 is now available to developers through the Gemini API and Google AI Studio. To facilitate adoption, Google is providing a developer Colab notebook containing examples of model configuration and prompting strategies tailored for embodied reasoning tasks. This resource allows developers to quickly integrate the new capabilities into their projects and begin experimenting with enhanced robotic autonomy. The introduction of this model underscores the growing importance of reasoning in robotics. As artificial intelligence evolves, the ability of machines to understand context and physical laws becomes essential for safe and effective deployment in dynamic environments. This update positions Google's technology as a key enabler for the future of autonomous systems, promising more sophisticated interactions between robots and the world around them.