LingBot-VLA

LingBot-VLA is a Vision-Language-Action (VLA) foundation model developed by Ant Group Robbyant, designed to enhance real-world robotic manipulation. Trained on approximately 20,000 hours of teleoperated bimanual data from nine dual-arm robot configurations, it demonstrates strong cross-morphology generalization and data-efficient post-training capabilities.

LingBot-VLA is a Vision-Language-Action (VLA) foundation model developed by Ant Group Robbyant, designed to enhance real-world robotic manipulation. Trained on approximately 20,000 hours of teleoperated bimanual data from nine dual-arm robot configurations, it demonstrates strong cross-morphology generalization and data-efficient post-training capabilities.

• 01Trained on 20,000 hours of real-world dual-arm robot data
• 02Evaluated on the GM-100 benchmark across three platforms
• 03High training throughput on commodity GPU clusters