Honor humanoid robot beats human world record in Beijing half marathon

by Liam O'Connor Sports Editor
Honor humanoid robot beats human world record in Beijing half marathon

A humanoid robot developed by Chinese smartphone maker Honor won the second Beijing E-Town Half Marathon on April 19, 2026, completing the 21-kilometer course in 50 minutes and 26 seconds — faster than the human world record set by Ugandan runner Jacob Kiplimo in Lisbon just weeks earlier.

The victory marked a dramatic turnaround from the inaugural race in 2025, when the winning robot finished in 2 hours, 40 minutes and 42 seconds and most entrants failed to complete the course. This year, over 100 teams participated — up from 20 — and nearly half of the robots navigated the course autonomously, a significant increase from last year’s reliance on remote control.

Honor’s winning robot, which too secured second and third places for its teammates, was designed with legs measuring 90 to 95 centimeters to emulate elite human runners and incorporated liquid cooling technology adapted from its smartphones. Du Xiaodi, an engineer on the Honor team, said the robot had been in development for a year and emphasized that advances in speed and endurance enable broader technology transfer to industrial applications like structural reliability and manufacturing.

Despite the triumph, the race was not without mishaps: one robot fell at the start line, another collided with a barrier, and technicians were seen attending to fallen units throughout the event. Beijing E-Town reported that about 40% of robots used autonomous navigation, while the remainder were remotely controlled. A separate remotely operated Honor robot crossed the finish line first in 48 minutes and 19 seconds, but the championship was awarded under weighted scoring rules that favored autonomous performance.

CCTV reported that the two runners-up, also from Honor and operating autonomously, finished in approximately 51 and 53 minutes. A humanoid robot was deployed as a traffic officer, using arm gestures and voice commands to guide participants — a symbolic nod to the blending of machine and human roles in the event.

Spectators noted the shift in attention from human runners to the machines. Sun Zhigang, who attended with his son, said he had never imagined robots surpassing humans in such a race. Wang Wen, another attendee, remarked that the robots’ speed “far exceeds that of humans” and may signal “the arrival of sort of a new era.”

Engineering student Chu Tianqi of Beijing University of Posts and Telecommunications praised the robots’ running posture, noting that despite AI’s short development timeline, the level of performance was impressive. Meanwhile, industry analysts at Omdia identified AGIBOT, Unitree Robotics, and UBTech Robotics Corp. As the only first-tier vendors in global shipments of general-purpose embodied intelligent robots, each having shipped over 1,000 units in the previous year, with the first two exceeding 5,000 units.

Key Detail The winning Honor robot’s time of 50:26 was not only faster than the human half-marathon world record but also represented an improvement of over 110 minutes compared to last year’s winning robot.

How did the robot beat the human world record if it ran on a separate course?

The robots and human runners competed on parallel tracks to avoid collisions, but the distance and course elevation were identical, allowing for a direct time comparison.

Why didn’t the remotely controlled robot that finished first in 48:19 win the race?

Although a remotely operated Honor robot crossed the finish line first, the championship was determined by weighted scoring rules that prioritized autonomous navigation, awarding the victory to the self-guided robot that finished in 50:26.

What does this race indicate about the current state of humanoid robotics in China?

The rapid progress — from failed attempts to world-beating times in one year — underscores China’s accelerated investment in embodied AI, with multiple companies now producing robots at scale and demonstrating real-world mobility and endurance capabilities.

You may also like

Leave a Comment