By RWT Robotics Editorial Team Published
CONFIRMED
A humanoid robot named Qestr lined up at Ultima Camp for the Antarctic Ice Ultra as one of 25 participants, Reuters reports, and organizer Oliver Wang called it the first humanoid ever to take part in the race. It did not finish. Reuters-derived coverage says Qestr covered about 3 km before leaving the course, and one report says its battery ran out.
What we know
The core facts come from Reuters. In an X post and a video report titled “Man vs. machine in Antarctica: Humanoid robot races with elite runners,” Reuters described Qestr joining runners in freezing conditions for a marathon in Antarctica. The setting was Ultima Camp, and Qestr was one of 25 participants.
A Reuters video report, syndicated by the Bangkok Post, quotes organizer Oliver Wang saying Qestr was the first humanoid robot to take part in the Antarctic Ice Ultra. That is a claim about the event’s own history, and the organizer is the right person to make it.
Then the result. Reuters-derived coverage aggregated by Ground News says Qestr covered about 3 km before leaving the race. The same body of coverage includes one report saying its battery ran out. That is the only cause we have seen attributed to a Reuters-based account, and it comes from secondary coverage rather than a statement from the organizer or the robot’s team.
We have not been able to read the original Reuters article text. Everything above is drawn from the Reuters video report, its X post, and coverage that syndicated or aggregated it. Dates also vary slightly between sites, with some placing the race on November 8, 2026. We are not treating any single date as settled.
What is rumored
A lower-quality running site, fatmarathoner.com, goes further than Reuters-derived coverage. It reports that Qestr’s lithium-ion battery failed despite a custom thermal jacket, with temperatures around -24°C. No official source has confirmed the jacket, the battery chemistry or the temperature figure, so we are treating all three as unverified.
The story is plausible. Cold is hard on batteries, and a thermal jacket is the kind of fix a team would try. But plausible is not the same as confirmed, and specific numbers from a weak source are exactly the details that drift as a story gets passed around.
What we don’t know yet
The biggest gap is control. None of the sources we could access says whether Qestr ran autonomously, was teleoperated, or followed a scripted gait with a human nearby. For a robotics story, that is not a footnote. A robot walking 3 km on ice under its own perception and planning is a very different achievement from a robot walking 3 km while someone steers it with a joystick.
We also do not know who built Qestr. No manufacturer, lab or startup is named in the coverage we reviewed. That means no spec sheet, no stated battery capacity, no published runtime, and no way to compare it with known platforms.
Human support is the third unknown, and it is the one our angle hinged on. We cannot tell you how many people accompanied Qestr, whether it was tethered or spotted, whether it was carried or recovered after stopping, or whether anyone swapped or warmed its battery during the attempt. The coverage simply does not say.
Finally, the course. Earlier editions of the Antarctic Ice Ultra have been run on a 4.2 km loop around camp for safety. If this year used the same layout, about 3 km would be short of a single lap. We have not seen confirmation of this year’s course, so that comparison stays conditional.
What it means
Humanoid robots have been showing up at human races for a while now. In April 2025, humanoids raced humans at the Beijing Yizhuang half-marathon, where Tiangong Ultra won the robot class in 2 h 40 min. In 2026, the Honor Robotics D1 reportedly finished the Beijing E-Town Humanoid Robot Half-Marathon in 50 min 26 s. Those were paved urban courses with support infrastructure close at hand.
Antarctica is a different test. Snow and ice change foot contact and traction. Cold changes what a battery can deliver. Recovery is harder when things go wrong. A robot that does well on asphalt in Beijing has not proven anything about a polar camp, and this attempt shows that gap is real.
Is 3 km a failure? Against the marathon distance, yes. As an engineering data point, it depends entirely on the unknowns above. If Qestr walked 3 km autonomously on ice in deep cold, that would be a meaningful result with a clear next problem to solve: energy in the cold. If it was teleoperated with a support crew close by, it is mostly a demonstration of a walking platform surviving briefly in harsh conditions.
The honest reading right now is narrow. Reuters confirms Qestr entered, the organizer confirms it was a first for this race, and Reuters-derived coverage confirms it stopped after about 3 km, with one report blaming the battery. Everything else, including the thermal jacket story and the question of who was really driving, still needs a primary source.
What would settle it is simple: the builder stepping forward with control mode, battery specs, runtime data and an account of what the support crew did. Until that happens, Qestr’s run is a good headline and an incomplete engineering story. We would rather tell you which parts are which than pretend otherwise.
Follow every development in one place on our humanoid robot tracker.
Keep reading
Robotics NewsNvidia says GR00T 2 learns from human video, but gives no release dateNvidia says GR00T 2 hits 71% on one demo per task. The figures are company claims, and no…Nov 9, 2026 · 4 min read
Robotics NewsBoston Dynamics puts wheels on Atlas to test whole-body reinforcement learningBoston Dynamics' 'Atlas On a Roll' post shows wheeled Atlas as a whole-body RL experiment. It is research,…Nov 9, 2026 · 4 min read
Robotics NewsNvidia expands Korea physical AI work with Samsung, Hyundai, Doosan and SK hynixNvidia named Samsung, Hyundai, Doosan and SK hynix in its Korea physical AI update. Stakes in four startups,…Nov 9, 2026 · 5 min readCredits and sources
By RWT Robotics Editorial Team
- Published
