Chinese AI firm Robbyant launched next-gen spatial perception models for robots, including LingBot-Depth 2.0 and LingBot-Vision. LingBot-Depth 2.0, built on the Masked Depth Modeling (MDM) technique, was trained on 150 million samples and achieved leading results across 12 of 16 depth completion benchmarks. The model reduces depth error by more than half compared with its predecessor in scenarios with severe depth loss, lowering the RMSE score from 0.132 to 0.062. It also improves detection of glass surfaces, mirrors, and other transparent objects. LingBot-Vision, a visual foundation model, is the first in the industry to use “boundary structure” as a pre-training objective, achieving sub-pixel-level boundary localization. It was trained on a relatively smaller dataset of 160 million images. For commercial deployment, LingBot-Depth 2.0 received certification from the Depth Vision Laboratory of Orbbec, with tests using chip-level depth data from Orbbec’s Gemini 330 series stereo 3D cameras showing improvements in edge detection, object contours, and performance under challenging lighting and material conditions. The collaboration will extend to new hardware solutions, including the RGB-D EGO device featuring a customized version of LingBot-Depth.