The AgiBot X2 Ultra hardware-software complex is an advanced new-generation humanoid platform designed for interactive service, show programs, telepresence, and robotics research. Standing approximately 1.31 m tall and weighing about 39 kg, the robot combines a friendly appearance with high mobility: up to 30 degrees of freedom, human-like gait, complex dance moves, and head mimicry synchronized with voice.

The X2 Ultra was initially designed as a "living interface" between humans and AI. It supports multimodal interaction: sight, voice, touch, and expression of "emotions" thanks to an interactive RGB camera, front and rear RGB cameras, a touch sensor in the head area, a microphone array, and an external wireless microphone. This allows the robot to recognize faces, gestures, objects, and environmental context, conduct natural dialogue, and work organically at events, showrooms, classrooms, and demonstration areas.

The X2 Ultra's advanced sensor system includes 3D LiDAR, an RGB-D camera, and a multi-line lidar for precise 3D-SLAM and depth perception. The robot is capable of autonomous navigation, route planning, proactive obstacle avoidance, and automatic recharging at a docking station. A battery with a capacity of approximately 500 Wh provides up to 2 hours of operation when walking at 0.5 m/s; direct charging, a replaceable battery, and an optional auto-charging station are supported. The maximum walking speed reaches 1.8 m/s, making the robot's movements dynamic and natural.

The Ultra version is equipped with an enhanced computing platform: two RK3588 boards and a high-performance Orin NX module (up to 157 TOPS), an extended set of interfaces (additional USB ports, LAN, video output, power for external modules), and support for 4G/5G communication. The open architecture and official support for secondary development allow for creating custom scenarios, integrating cloud services, enterprise systems, and proprietary AgiBot manipulators, transforming the X2 Ultra into a flexible platform for commercial and research projects.

Multimodal interaction: sees, hears, feels touch



The X2 supports visual, voice, tactile interaction, and facial "mimicry," providing a natural communication format with people. An interactive RGB camera, head touch sensor, microphone array, and wireless microphone make dialogue and show scenarios as lively as possible.

Lively gait, dances, and synchronized head movements



The robot supports humanoid gait and complex dance sequences with head movements synchronized with voice. 25–30 degrees of freedom, including the head and waist, provide expressive plasticity and spectacular performances on stage, at a booth, or at a company stand.

Powerful computing platform and readiness for customization



Inside the X2 Ultra are two RK3588 boards and an additional Orin NX module up to 157 TOPS, as well as extended interfaces: more USB, two LAN, video output, power lines for external modules, and 4G/5G communication. Secondary development is officially supported, so integrators can add their services, AI models, and business logic.