Gemini Robotics 1.5 is Google’s most advanced vision-language-action (VLA) model, allowing robots to interpret visual information and natural language instructions into precise motor commands. Unlike conventional robots, this model doesn’t just act—it thinks before taking action, evaluating tasks step by step and even explaining its reasoning. This level of transparency helps developers and users understand how robots make decisions, enhancing trust in intelligent automation.

With Gemini Robotics 1.5, robots can handle complex tasks such as sorting objects, managing inventory, or tidying environments, all while adapting to new conditions and learning across different physical embodiments. Its multimodal understanding accelerates skill acquisition, enabling robots to generalize learned actions to diverse situations efficiently.

Gemini Robotics-ER 1.5

Gemini Robotics-ER 1.5 serves as the high-level brain for these robots, orchestrating activities and creating detailed, multi-step plans. This AI reasoning model excels at spatial understanding, tool usage, and decision-making in physical environments. It can natively call digital tools such as Google Search to gather real-time information, plan steps for a task, and execute actions with precision.

Together, Gemini Robotics 1.5 and ER 1.5 enable robots to perform tasks that require contextual reasoning, complex planning, and real-world adaptability, making them ideal for industries ranging from manufacturing and logistics to home automation and service robotics.

Starting today, Gemini Robotics-ER 1.5 is available via the Gemini API in Google AI Studio, while Gemini Robotics 1.5 is accessible to select partners. These models provide developers with the tools to build more capable, versatile robots that can perceive, plan, think, and act autonomously, helping accelerate the next generation of AI-driven physical agents.

As AI continues to merge with robotics, Gemini Robotics 1.5 represents a significant leap toward intelligent robots capable of real-world problem-solving, ushering in an era where AI-powered automation can efficiently handle complex, multi-step tasks across diverse environments.