One Robot, Multiple Functions，Intelligence Across Scenarios
Top intelligence, deeply integrated into diverse
scenarios, unlocking infinite potential
MoNet large model
Autonomous thinking, multimodal understanding multitasking, zero-shot millisecond-level inference
Model parameter quantity
6B
Terminal-side deployment
Mixture of Experts
Inference speed
5MS/TOKEN
Semantic intelligence
Combine dynamic modeling
semantic understanding、future prediction
Semantic type
100kinds
Detection accuracy
5cm
Running time
20ms
Precision equipment
LiDAR + camera
Spatial intelligence
Develop zero-shot transfer
multimodal decision-making technologies
enable centimeter-level accurate navigation
Area of autonomous mapping
100,000㎡
Accurate navigation detection accuracy
±5cm
Autonomous obstacle avoidance
recognition height
≥10cm
Microphone array: 6
Provide clear and accurate voice recognition
360° directional pickup
360° directional pickup
RGB depth camera + panoramic camera
Accurately perceive faces and environmental objects
3D LiDAR
Realize 3D environmental perception and
autonomous path planning
autonomous path planning
AI Large Language Model
Supports custom LLM API integration
custom RAG knowledge bases
custom RAG knowledge bases
Full-body degrees of freedom
40
Height
167cm
Body weight
70kg
RGB depth camera +
panoramic camera
panoramic camera
Perceive environmental objects
and distances
and distances
