First Ant Lingbo Embodied Large Model Challenge Launches at Bund Conference

Shen Yujun, chief scientist of Ant Lingbo Technology; Huang Yongtao, chief data scientist; Shi Hongzhu, operations head of the ModelScope Community; Wang Hongqiang, director of AMD AI software products and developer ecosystem; and Cui Yuanzheng, head of the Alibaba Cloud Tianchi platform, jointly opened the contest. Organizers said the contest aims to provide a clear practical path for developers to work directly with LingBot-VLA, understand it through use, improve it in tasks, and expand the model's applications with their own devices, scenarios and ideas.

LingBot-VLA is an open-source embodied intelligence foundation model from Ant Lingbo. It was first released in January 2026, and LingBot-VLA 2.0 was released in July of the same year, with the model and post-training code fully open-sourced to the community. Jiang Bo, head of market operations and technical ecosystem at Ant Lingbo, said at the forum: "For an open-source model, release is only the starting point. The model's real value lies in whether developers are willing to use it, whether they can train, fine-tune and adapt based on it, and whether they can apply it to different robots and tasks."

The competition has an online preliminary round and an offline final. Registration and the online preliminary stage will run until Oct. 26. The preliminary round includes a mandatory RoboTwin 2.0 simulation task and an optional real-machine innovation task, and participating teams can apply for dedicated cloud computing support under the rules. Finalist teams will attend an offline real-robot hackathon in Shanghai from Nov. 13 to 15, where they will continue model tuning and real-machine deployment using real robot platforms, datasets and computing power provided by the organizers.

Through the challenge, Ant Lingbo aims to bring LingBot-VLA to a broader developer community and university research community, so that more developers can use their own tasks, methods and ideas to expand the model's possibilities. The company also aims to connect developers, robot platforms and real-world scenarios around the open-source base, giving more embodied intelligence ideas a chance to move from experiment to practice and supporting open collaboration and scenario innovation.

The registration link is https://tianchi.aliyun.com/competition/entrance/532514. The article was provided by Ant Lingbo and republished by QbitAI with authorization; the views belong to the original author.