🧔♂️ A friendly human may check it before it goes live. More news here
China tests humanoid robots in half-marathon
More than 300 humanoid robots from over 70 teams will run Beijing’s second robot half-marathon on April 19, tackling a 21 km course with paved slopes and parkland as China uses the event to test advances in durability, battery life, and autonomous movement.
Organizers said nearly 40% of entrants will navigate without remote control, up from none last year, when several robots fell near the start and the winning Tiangong Ultra finished in 2 hours and 40 minutes.
Counterpoint Research said China accounted for over 80% of 16,000 humanoid robots installed worldwide in 2025, but analysts and founders said the industry is still at an early stage and far from broad industrial or home use because software, dexterity, and cost remain major hurdles.
🔗 Source: Reuters
🧠 Food for thought
Implications, context, and why it matters.
China’s robot makers sell thousands of units, but mainly for research, not factory work
- Events like the marathon test basic abilities, yet some Chinese humanoid robot firms already ship products in volume 1.
- Unitree, a Chinese robotics company, delivered more than 5,500 humanoid robots in 2025. UBTECH, another Chinese humanoid robot maker, increased revenue from its full-size humanoids by more than 22 times that year 12.
- Factory adoption still lags. Research and education buyers accounted for over 70% of Unitree’s 2025 humanoid sales 3.
- Industrial rollouts remain small, with UBTECH robots mostly running pilot programs at automotive plants for brands such as Audi and Zeekr, a Chinese electric vehicle brand 4.
High profit margins and Nvidia dependency create a new pattern for hardware
- Work to make robots tougher is reshaping how these companies earn money, with some posting strong profits 1.
- Unitree said its humanoid robots reached a gross margin above 60% in 2025. Tesla’s vehicle margin ran around 18% in the same period 1.
- Companies get there by bringing core parts like motors and sensors in-house, plus tight cost control 5.
- AI features still lean on Nvidia, the US chip company whose processors power much of today’s AI computing 3.
- That dependence on overseas chips turns robotics into a flashpoint in the tech supply chain and adds pressure for China to build a stronger domestic semiconductor industry 3.
Stay updated on the go with our mobile app.
Get latest insights with smoother, more personalized experience through TIA mobile app.




