Robbyant Releases LingBot-VLA 2.0: An Open-Source 6B Vision-Language-Action (VLA) Model for Cross-Embodiment Robot Manipulation

A Vision-Language-Action model is an AI system that takes camera images and language instructions as input and outputs commands to control a robot. LingBot-VLA 2.0 is an open-source version of this technology that works across different robot types by using a unified representation of actions and body parts. The model addresses a practical gap where robot AI systems often perform well in laboratory settings but fail when deployed in real environments, advancing generalization, action space, and predictive capabilities through improved training data curation and architecture design. The system processes robot trajectories and human videos to learn manipulation tasks and can run inference in about 130 milliseconds on standard hardware.