# Robot contest: team TARS, named after a film robot, wins the 48-hour final

> Teams from universities, labs and companies had to get real robot arms to pick up fruit, cosmetics and waste, all driven by the same starting AI model. According to the Chinese site Leiphone, the final of the Ant Lingbo contest lasted 48 hours.

Published on 2026-11-17

According to Leiphone, team TARS, whose average age is 25, scored 82.5 points and took first place. Its name comes from a loyal, funny robot in a film, the site says.

The finalists learned the tasks on the morning of the opening, then had 48 hours to put their tuning to the test on real robots.

All the teams started from the same open model, LingBot-VLA 2.0, with the same hardware. There were three tasks: picking up fruit and vegetables with one arm, putting cosmetics into a transparent box, then sweeping up waste with two arms and tipping it into a basket.

According to Leiphone, captain Wu Haonan, who is about to graduate with a master's from Tianjin University, set out his strategy: reduce a cluttered scene to the picking up of a single object, to get around the model's perception limits.

The site says 1,855 teams and 3,252 participants from 13 countries signed up, from 648 institutions and companies. It describes the contest as the largest in the field for this type of model.

## Sources

- [雷峰网 Leiphone](https://www.leiphone.com/category/industrynews/LTOeijdeWR08fZGP.html)

---

HTML version: https://vuntum.com/en/news/team-tars-average-age-25-wins-robot-contest-with-82-5-points

Factual data: CC BY-NC 4.0. Attribution: “Source: Vuntum (https://vuntum.com)” with a link to the page quoted. Editorial text: all rights reserved, short quotation with attribution.
