Honor robot runs half marathon in 50:26, seven minutes faster than the world record

An autonomous humanoid robot from Honor ran a half marathon (21.1 km) in 50 minutes 26 seconds, 7 minutes faster than the current human world record. The run took place in China, and the robot operated entirely on its own, with no operator involved. It is the first officially recorded case of a humanoid robot beating an elite human athlete over a classic running distance. Honor is widely known as a smartphone maker.

AI-processed from Wired; edited by Hamidun News
Honor robot runs half marathon in 50:26, seven minutes faster than the world record
Source: Wired. Collage: Hamidun News.
Listen to article

A humanoid robot from Honor ran a half marathon in 50 minutes 26 seconds, beating the human world record by more than seven minutes. The run took place in China and is the first officially recorded case of a fully autonomous anthropomorphic robot outrunning the best human athlete over a classic running distance.

A half marathon is 21.097 kilometers. The men's world record belongs to Ugandan runner Jacob Kiplimo: 57 minutes 31 seconds, set in November 2021 in Lisbon. Honor's result of 50:26 means a gap of almost seven minutes.

For scale: in professional running, a few seconds separate first and fifth place at a major race. A seven-minute margin over the absolute world record is not a race win but a jump into a different category of speed and endurance.

The key detail is that the robot acted fully autonomously throughout the run. There was no remote control. Its algorithms regulated pace, maintained dynamic balance and made decisions at every step without an operator. That is what sets Honor's result apart from the teleoperated demonstrations and show runs with significant technological limitations that were staged before.

Attempts to put robots on a track have been made before. Research groups have shown bipedal robots that can run, but either over short distances, in laboratory conditions, or with limited autonomy. A full half marathon in autonomous mode is a qualitatively different level of task.

Honor is widely known as a smartphone maker. The company spun out of the Huawei ecosystem in 2020 after US sanctions and grew into an independent consumer electronics player with a notable market share in China and Southeast Asia. Entering humanoid robotics, and doing so immediately with a record result, looks like a deliberate statement of technological expansion beyond its usual segment.

The achievement fits into a large state-backed context. China is systematically increasing investment in bipedal robotics. The country's Ministry of Industry and Information Technology has named humanoid robots a strategic priority and set the goal of establishing their industrial production by 2025. Dozens of companies, including Unitree, AgiBot, Fourier Intelligence and Deep Robotics, compete for leadership in the segment. In the West, Boston Dynamics, Figure AI, Agility Robotics and Tesla with its Optimus robot are working the same niche.

The task is exceptionally hard technically. Bipedal locomotion is unstable by nature: a human is essentially falling forward continuously and catching themselves in time. In running, the impact load on joints rises many times over compared with walking, and compensating for surface irregularities requires micro-corrections within fractions of a second. Holding an average pace of about 2 minutes 24 seconds per kilometer for 50 minutes without a single failure is an engineering result that until recently seemed achievable only in the long term.

The practical significance of the record goes beyond a sporting event. Robots that can work reliably under intense physical load for an hour become usable in real scenarios: warehouse logistics, emergency response, industrial environments with unstructured terrain, perimeter security. Honor's record is not a hype stunt but a public verification of the level of reliability needed for commercial deployment of humanoid robots.

The humanoid robot race has become measurable, in seconds and kilometers.

Story
ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Unified AI Workspace · Nomad

Test Frontier LLMs & Autonomous Agents in One Workspace

Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.

Try Nomad AI →Launch Nomad Agent
What do you think?
Loading comments…