Like Astribot S1, the T1 uses cable-driven motion for smoother, more natural robot movement.

Chinese robotics startup Astribot, also known as Stardust Intelligence, has unveiled one of China’s most affordable humanoid robots to date.
The company has officially opened orders for the T1, a compact humanoid model priced from around $13,000, significantly lower than its flagship S1 robot, which costs nearly $100,000.
The compact wheeled humanoid features 23 degrees of freedom excluding end effectors, with a single-arm payload capacity of 11 pounds (5 kilograms).
The launch reflects China’s growing push to accelerate the commercialization of lower-cost humanoid robots for wider real-world adoption.
Last year, the Shenzhen-based firm launched Astribot S1, a humanoid assistant designed to handle household chores, pet care, exercise, music, and kitchen tasks with advanced agility, dexterity, and precision.
Low-cost robot platform
Astribot is a robotics startup founded in 2022, and since 2024, the company has reportedly raised around $100 million to expand development of its AI-powered humanoid systems. Its latest humanoid robot, T1, aims to be the most accessible developed in China so far.
Designed for practical real-world deployment, the T1 combines customizable AI computing, human-like dexterity, and flexible application support across multiple industries. The humanoid stands 155 cm (5 feet 1 inch) tall, weighs approximately 66 kg (146 lbs), and includes 23 degrees of freedom excluding end effectors. It can carry payloads of up to 5 kg (11 lbs) with one arm, enabling it to perform a wide variety of operational tasks.
The robot supports robotic grippers and five-fingered end effectors for delicate manipulation. According to the company, the T1 can assist with cooking, laboratory operations, laundry folding, automotive parts sorting, and EV charging tasks. Like the larger S1 model, the T1 uses a cable-driven motion architecture, enabling smoother, more natural movements than traditional robotic systems.
Astribot says the humanoid’s AI is trained primarily through human demonstration data, helping it learn real-world behaviors and workflows more efficiently. The company is promoting the T1 as a practical platform for the testing, development, and commercialization of next-generation humanoid robots.
Advanced robot learning
The Astribot S1 is a dual-arm mobile humanoid robot developed by Astribot, designed for full autonomy rather than relying on teleoperation like many other humanoid systems.
Its core capability is a whole-body imitation learning system that enables it to learn tasks by observing human demonstrations and executing them independently with high precision. The robot first appeared at the 2024 World Robot Conference in Beijing, followed by continued refinement detailed in a July 2025 arXiv paper describing the “Astribot Suite.”
Technically, the S1 stands about 170 cm tall and weighs around 198 pounds (90 kilograms). It features 23 degrees of freedom, including 7-DoF per arm, a 4-DoF torso, a 2-DoF head, and a 3-DoF omnidirectional base. Its cable-driven architecture mimics human musculature, offering smooth motion, high payload efficiency, and safer human interaction. Each arm can lift to 11 pounds (5 kilograms), while advanced sensors include RGB-D cameras, LiDAR, IMUs, force and tactile sensors, and more.
Powered by the Astribot Suite, the system combines VR-based data collection, imitation learning policies, and real-time trajectory optimization, achieving task success rates of 80–100 percent. Integrated language models further enable natural interaction and instruction-based control.
Recommended Articles
Get the latest in engineering, tech, space & science - delivered daily to your inbox.
Jijo is an automotive and business journalist based in India. Armed with a BA in History (Honors) from St. Stephen's College, Delhi University, and a PG diploma in Journalism from the Indian Institute of Mass Communication, Delhi, he has worked for news agencies, national newspapers, and automotive magazines. In his spare time, he likes to go off-roading, engage in political discourse, travel, and teach languages.






















