Ant Lingbo Releases LingBot-Vision & LingBot-Depth2.0 Robot Vision Models

量子位 · wechat · 2026-07-07

Ant Lingbo has released LingBot-Vision, a spatially native vision foundation model for robotics, alongside the depth perception model LingBot-Depth2.0. These models focus on improving issues like fragmented and hollow depth maps in scenarios involving transparent or reflective materials, small objects, long distances, and low-light occlusions. This provides robots with a more stable spatial framework for grasping and navigation, an achievement hailed by netizens as a major milestone in embodied vision.

Related event: Ant Robbyant Open-Sources LingBot Vision Models, Topping Depth Benchmarks(18 posts)→

Original post →

More from Embodied

Embodied channel →