Krafton’s Ludo Robotics unveiled Ludi 0.2, extending a system that combines a vision-language model with navigation and manipulation. The company says the update improves responses to human feedback and social signals, task memory and reasoning, gesture understanding, and coordination between tools and actions. It also targets smoother motion in defined environments. Ludi 1.0, intended to integrate perception, conversation and physical action into one model, remains under development. The report does not provide independent performance comparisons or success rates across diverse real-world settings.