LingBot-VLA 2.0
An embodied intelligence foundation model developed by Ant Lingbo, trained from scratch for the physical world and adapted to multiple robot configurations during pretraining. Its core feature is "one brain, multiple machines": a single model can drive bodies of different configurations, reducing the cost of transfer across scenarios and hardware.
Timeline (last 90 days)
Ant Lingbo's self-developed LingBot-VLA 2.0 foundation model has already adapted to 20 robot configurations from 17 brands during the pre-training stage.
At the Industry Fair site, two robot bodies with different configurations from Tihu and Stardust Intelligence were demonstrated, both equipped with the Lingbo brain LingBot-VLA 2.0, completing high-precision industrial loading and unloading and flexible sorting tasks respectively.
All Facts (2)
Ant Lingbo's self-developed LingBot-VLA 2.0 foundation model has already adapted to 20 robot configurations from 17 brands during the pre-training stage.
50%UnverifiedAt the Industry Fair site, two robot bodies with different configurations from Tihu and Stardust Intelligence were demonstrated, both equipped with the Lingbo brain LingBot-VLA 2.0, completing high-precision industrial loading and unloading and flexible sorting tasks respectively.
50%