Gemini Robotics 1.5

Gemini Robotics 1.5 is Google DeepMind's multitask VLA model extending Gemini-LLM capabilities into the physical world. Supports cross-embodiment zero-shot generalization. In July 2026, Google DeepMind launched Gemini Robotics Early Access program with ER 1.6 (Embodied Reasoning v1.6), adding a waitlist for developers and demonstrating precise bimanual manipulation capabilities.