The second World Humanoid Robot Games concluded in Beijing on Wednesday after five days of competition, with Tiangong Ultra providing one of the highlights by winning the large-size robot 100-meter race in a Games-record 8.64 seconds.

Developed by the Beijing Humanoid Robot Innovation Centre, Tiangong Ultra lowered its time repeatedly over the course of the Games. It clocked 9.39 seconds earlier in the competition before improving to 8.86 and then 8.64 in the final held on the last day.

The closing ceremony was held at the National Speed Skating Oval following the final day of competition. A total of 2,056 humanoid robots representing 666 teams from around the world competed across 51 events, with 1,301 contests staged over the five days.

Several events produced major improvements from the inaugural edition. The large-size robot 400m record fell from 1:28.03 last year to 38.15 seconds, while the best 1,500m time improved from 6:34.40 to 2:21.64. The standing high jump record also rose from 0.95641 meters to 3.40 meters.

The program combined traditional sporting contests, while testing the robots' ability to operate in real-world environments. Robots competed in football, table tennis, gymnastics, weightlifting, martial arts, tug-of-war and combat events, while 21 scenario-based events included tasks related to factories, hotels, supermarkets and emergency response.

The Games also served as a testing ground for embodied artificial intelligence, allowing developers to assess robots' mobility, autonomous decision-making and ability to cope with unpredictable environments.

At the closing ceremony, organizers released a dataset containing more than 2,500 hours of real-world operational data gathered from training and competition. The dataset covers 12 application scenarios, more than 100 skills and over 10,000 detailed tasks, and will be made available free of charge to companies, universities and research institutions.

Organizers also announced that the third World Humanoid Robot Games will be held in Beijing in August 2027, with scenario-based events expected to make up a larger part of the next program.