Google DeepMind has introduced Gemini Robotics 2.5, an advanced vision-language-action model designed to enhance robot learning and dexterity in real-world scenarios. This new model significantly improves a robot's ability to perform a wider range of tasks with greater precision and speed, achieving success rates 30% to 100% higher than previous versions. In a practical application, Gemini Robotics 2.5 powered Apptronik's Apollo 2 humanoid robot to successfully pick and pack automotive parts on Mercedes-Benz production lines with a 96% success rate, demonstrating its potential for industrial use. AI
IMPACT Enhances robot capabilities for real-world tasks, potentially accelerating adoption in manufacturing and logistics.
RANK_REASON New model release from a frontier AI lab (Google DeepMind) with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →