Google DeepMind introduced Gemini Robotics 1.5 and Gemini Robotics-ER 1.5 on 25 September 2025. The vision-language-action model connects visual information and instructions with motion, while the ER model supports environmental understanding and reasoning. The announcement also described developer access to ER through the Gemini API.
Understanding “move the red object” is only one part of the system. The gripper must still hold it, sensors must be calibrated, and motion must remain within the application’s limits. Model access does not make every robot compatible.
A useful first experiment is sorting a limited range of lightweight objects. Send uncertain detections to an operator and test with the customer’s own objects. Evaluate the camera, end effector, compute requirements and integration together.
Original sources
Source summaries and application analysis by COCON. Pilot ideas are proposals, not claims of completed COCON projects or confirmed product availability.
Could this work in your operation?
Share your workflow, location and target outcome. Start with a feasibility assessment, then validate the technology and define a pilot.
Discuss your project ↗ Email our team ↗
