Unitree’s UnifoLM-X2-1.0 reads the opponent, predicts the next move and reacts instantly. For the first time a world model controls a fully autonomous humanoid fight in real time.
Summary
UnifoLM-X2-1.0 is a world model that enables real-time autonomous combat for humanoid robots, breaking through bottlenecks in planning and decision-making and validating the feasibility of large-scale deployment.
Similar Articles
Unitree just dropped UnifoLM-WLA-1.0 — a single 6B model that does 64 whole-body + tabletop tasks on a real humanoid
Unitree released UnifoLM-WLA-1.0, a 6B-parameter whole-body vision-language-action foundation model trained on ~2,500 hours of real robot data that handles 64 tasks on the Unitree G1 humanoid, using an MMDiT action expert on top of an embodied reasoner based on Qwen3-VL.
Real-Time Omni-Modal Interaction Driven Whole-Body Mobile Manipulation
Unitree releases UnifoLM-OminiA-0.3, a single AI model for whole-body mobile manipulation with real-time omni-modal interaction, enabling autonomous home-care and wellness tasks.
Unitree unveils its new H2-X humanoid by having it perform kung fu kicks powerful enough to dent car doors. The speed, precision, and sheer force of those movements are insane.
Unitree unveiled its new H2-X humanoid robot by showcasing it performing powerful kung fu kicks capable of denting car doors, highlighting impressive speed, precision, and force.
Unitree G1 and EngineAI PM01 fight
This article covers a simulated or staged fight between Unitree's G1 humanoid robot and EngineAI's PM01 robot.
UNITAS: A 3D-Native World Action Model for Embodied Manipulation
UNITAS是首个3D原生的世界动作模型(WAM),在共享的度量3D坐标系中统一观测、动作与场景动态,以1.7B参数在RoboTwin场景预测上取得最佳结果(位移误差较PointWorld降低最多49%),并在LIBERO上达到99.8%的成功率。