Ant Group Upgrades LingBot-VLA 2.0
jiqizhixin · x · 2026-07-16
Ant Group has released LingBot-VLA 2.0, aiming to transition VLA robots from the lab to real-world applications.
Core approaches include:
- Trained on 60,000 hours of diverse data, comprising roughly 50,000 hours of robot data and 10,000 hours of human videos.
- Enabled the robot to control its head, waist, chassis, and dexterous hands, enhancing multi-part coordinated manipulation.
- Introduced predictive dynamics to improve temporal reasoning capabilities.
The team claims the new version outperforms its predecessor in the general settings of the GM-100 benchmark, demonstrating strong cross-embodiment mobile manipulation skills across two robotic platforms. The post also includes links to the paper, project, code, and a report by Jiqizhixin.
Related event: Ant Group unveils LingBot-VLA 2.0 for cross-embodiment robot control(8 posts)→
More from Embodied
- NVIDIA shows 22 SIGGRAPH papers and Omniverse tools for robot simulation — facontidavide · 2026-07-22
- Nothing phone mockup turns a film joke into a modular design meme — ZeYanjie · 2026-07-22
- Lightwheel AI Launches SimReadyGen: Text-to-Physics-Accurate Robot Sim Assets — ZeYanjie · 2026-07-22
- Humanoid robot sorting packages in a warehouse sparks debate over job loss — MonaJalal_ · 2026-07-22
- NVIDIA pushes OpenUSD as the common layer for simulation and physical AI — MonaJalal_ · 2026-07-22
- A quadruped robot gets a custom glow-up with a new shell and screen — DynamicWebPaige · 2026-07-22