In a nutshell: Gemini Robotics-ER 1.6 enhances spatial understanding and enables robots to read complex instruments for the first time through specialized embodied reasoning training.
Google has released Gemini Robotics-ER 1.6, a language model specifically developed for robotics tasks that improves capabilities such as spatial understanding, instrument reading, and success detection. The model is intended to enable robots to perceive physical environments more precisely and autonomously execute complex tasks.
Gemini Robotics-ER 1.6 is the latest version of a model specifically designed for the requirements of autonomous robotic systems. Unlike general-purpose language models, this model focuses on “embodied reasoning” — the ability to connect digital intelligence with physical actions in the real world. The model addresses core robotics challenges such as visual perception, spatial understanding, task planning, and recognition of action success or failure.
Version 1.6 brings several concrete improvements compared to its predecessor Robotics-ER 1.5 and the general-purpose model Gemini 3.0 Flash. A new feature is the ability to read instruments — the model can now interpret complex pressure gauges and sight glasses. This capability emerged from collaboration with Boston Dynamics and addresses a practical need in industrial robot deployments. The model functions as a high-level reasoning engine and can invoke external tools, such as Google Search, Vision-Language-Action models (VLAs), or user-defined functions, to execute tasks.
Practitioners can now use Gemini Robotics-ER 1.6 via the Gemini API and Google AI Studio. Google additionally provides a developer Colab notebook that includes examples for configuring the model and formulating prompts for embodied reasoning tasks. Benchmarks show performance gains in spatial and physical understanding, particularly for tasks such as pointing, counting, and success detection.
Source: ainews-dev.lumi-systems.io · Published May 17, 2026
Lumi AI News — AI-assisted curation pursuant to Art. 50 EU AI Act. Paraphrase and classification by Lumi News Pipeline v1.5.2.