ChainThink reports that on April 14, according to 1M AI News, Google DeepMind released Gemini Robotics-ER 1.6, positioned as a high-level reasoning model for robotics. Compared to its predecessors ER 1.5 and Gemini 3.0 Flash, this model shows significant improvements in spatial reasoning and multi-view understanding, and is now available to developers via the Gemini API and Google AI Studio.
The core upgrades include three key capabilities: First, improved pointing accuracy enables precise object detection, counting, spatial relationship reasoning, and motion trajectory planning, while correctly rejecting references to objects not present in the scene. Second, multi-view successful detection allows the system to assess task completion by integrating inputs from multiple cameras, maintaining accuracy even in occluded or dynamic environments. Third, a new instrument reading capability has been added, enabling interpretation of industrial instruments such as circular pressure gauges, vertical level indicators, and digital displays through agentic vision with step-by-step reasoning.
The dashboard reading capability stems from the collaboration between DeepMind and Boston Dynamics. On the same day, Boston Dynamics announced that it has integrated Gemini and Gemini Robotics-ER 1.6 into the Orbit AIVI-Learning product, which went live for all AIVI-Learning customers on April 8. With the integration, enhanced dashboard support has been added, enabling the quadruped robot Spot to autonomously patrol industrial facilities and read data from instruments such as pressure gauges.
Boston Dynamics stated that, with Gemini's reasoning capabilities, AIVI-Learning has improved its baseline performance and accuracy on tasks such as visual inspection, pallet counting, and liquid detection. DeepMind said that ER 1.6 is its "safest robot model," demonstrating better adherence to safety instructions than ER 1.5 in adversarial spatial reasoning tasks; in safety risk identification tests based on real injury reports, the ER series models outperformed Gemini 3.0 Flash by 6% in text scenarios and 10% in video scenarios.