Vega
Production-ready humanoid with 36 DOF, 15 kg-payload arms and 20 h runtime—powered by Intel x86 + Jetson Orin and an open ROS/Python stack.
Humanoid.Guide skill score: 2/10 This score is calculated as the combined total of Navigation and Manipulation performance.

Specifications and details:
Availability | Production |
---|---|
Nationality | UK |
Website | http://dexmate.ai |
Degrees of freedom, overall | 36 |
Degrees of freedom, hands | 12 |
Height [cm] | 171 |
Manipulation performance | 1 |
Navigation performance | 1 |
Max speed (km/h) | 4 |
Strength [kg] | 15 |
Weight [kg] | 135 |
Runtime pr charge (hours) | 20 |
Safe with humans | Yes |
CPU/GPU | Intel x86, Nvidia Jetson |
Ingress protection | Not specified |
Camera resolution | Not specified |
Connectivity | Bluetooth, Ethernet, USB, WiFi |
Operating system | Linux |
LLM integration | Yes |
Latency glass to action | Not specified |
Motor tech | Brushless Motor |
Gear tech | Harmonic Gear reduction |
Main structural material | Aluminium |
Number of fingers | 10 |
Main market | Industrial automation, R&D |
Description
Dexmate’s Vega is a production-ready, US-built humanoid designed to bring adaptable automation to industrial floors and R & D labs alike. Standing 171 cm tall and walking at 4 km/h, the 135 kg robot moves on an omni-directional mobile base, then unfolds its torso to reach as high as 2.2 m for overhead tasks. A brushless-motor / harmonic-gear drive-train keeps motion smooth and quiet, while built-in safety logic lets Vega operate side-by-side with people. dexmate.ai
Power and dexterity are central to the platform. Each 7-DOF arm lifts 15 kg and ends in a five-finger hand (12 DOF total, 10 fingers) for fine manipulation; overall, Vega offers 36 degrees of freedom to match the mobility of a skilled worker. A high-capacity battery supports up to 20 hours of runtime per charge (manufacturer spec), so the robot can cover an entire shift without a pit stop. Robotic Gizmos
Edge AI comes from an Intel x86 CPU paired with an NVIDIA Jetson Orin GPU, running Linux plus Dexmate’s ROS-compatible SDK. Built-in RGB-D cameras, LiDAR, IMUs, ultrasonic sensors and force/torque sensing feed real-time perception, while optional LLM integration enables natural-language tasking and fast skill re-authoring. Developers get a Python API, URDF/USD files for leading simulators and rich I/O (Ethernet, Wi-Fi, USB, Bluetooth, DisplayPort, 5 V/12 V rails) for custom payloads. dexmate.ai
With a list price of ≈ $80 k and lead times under four months, Vega lowers the barrier to full-scale humanoid automation. Its long endurance, heavy-lift arms and open software stack make it a practical choice for assembly, inspection, logistics and advanced research—all while helping organizations tackle labor shortages and future-proof their operations.