Ant Lingbo

Ant Lingbo, Ant Group's embodied AI unit, signed an MOU with the Arab Federation for Digital Economy at a China-Arab matchmaking session during the fifth Global Digital Trade Expo. The partners will establish a standing cooperation mechanism to deploy embodied AI and full-stack robot solutions in the UAE, GCC and wider Middle East, covering hospitality, retail, logistics, industrial operations, smart city and healthcare, with Arabic-language localization and ecosystem collaboration. Contract value, unit volume and deployment timeline were not disclosed.

The first Ant Lingbo embodied AI large model challenge opened at the Bund Summit on September 14, co-organized with ModelScope and Alibaba Cloud Tianchi. It uses LingBot-VLA 2.0, a pre-trained model fully open-sourced in July 2026, as its technical base and is open to enterprises, universities, and individual developers worldwide.

The online preliminary round runs through October 26, with a required RoboTwin 2.0 simulation task, an optional real-robot innovation task, and dedicated cloud compute support. Finalists will join an offline real-robot hackathon in Shanghai from November 13 to 15.

At the 2026 Inclusion·Bund Conference, Zhu Xing, CEO of Ant Group’s embodied AI startup Ant Lingbo, said the company has begun its first independent fundraising round with a target of RMB 1.5B (about $214M). The raise is not for lack of money or computing cards, he said, but to improve incentives for existing and future senior talent and strengthen governance as an independent company. Ant Lingbo aims to close a second round by the end of the year, an unusually fast pace for an embodied AI startup.

Ant Lingbo has open-sourced a 1.3B-parameter lightweight LingBot-World 2.0 world model, following the 14B version released in July. It runs real-time world generation on a single consumer GPU. Unlike a pruned checkpoint, it was trained end-to-end with causal pretraining using MoBA attention masks, consistency distillation, and DMD distribution matching distillation.

The release includes the causal pretraining base and a bidirectional teacher for community post-training, distillation, and scene adaptation. AMD senior VP Jack Huynh demonstrated LingBot-World 2.0 during his IFA Berlin opening keynote.

Ant Lingbo (蚂蚁灵波) showed its LingBot-VLA 2.0 embodied general brain at the 2026 Bund Conference. The company says the same brain can drive robots from Leju Robotics and TiHu Robotics, among others, with pretraining already adapted to 17 robot brands and 20 configurations, covering single-arm, dual-arm, bipedal and wheeled bodies.

A deployment with Shanghai Guoda Pharmacy is live, Ant Lingbo says: robots take orders and pick medicines autonomously in aisles about 80 cm wide despite transparent, reflective packaging, and assist pharmacists with sorting at night, with no store renovation. Ant Lingbo also opened an embodied large-model challenge with ModelScope and Alibaba Cloud Tianchi.

Maniformer, a one-stop physical AI data service platform, has raised hundreds of millions of yuan (tens of millions of dollars) led by China Telecom. Zhangjiang Group joined, and existing investors Sequoia Capital China and Yuanqi Innovation made oversubscribed additions. Funds will scale mass production of MEgo no-body acquisition products, strengthen data-stack capabilities, and accelerate tens of millions of hours of physical interaction data. The company has worked with Ant Lingbo to supply training data for the LingBot-VLA 2.0 model.