@rohanpaul_ai: A 1B-parameter vision model just beat a 7B one on depth, frozen, single linear layer, zero fine-tuning. @robbyant_brain…
Summary
Robbyant releases LingBot-Vision, a 1B-parameter vision model trained on boundaries that achieves better depth estimation than DINOv3-7B, with open weights.
View Cached Full Text
Cached at: 07/09/26, 07:59 AM
A 1B-parameter vision model just beat a 7B one on depth, frozen, single linear layer, zero fine-tuning.
@robbyant_brain’s LingBot-Vision: 0.296 NYU-Depth v2 RMSE vs DINOv3-7B’s 0.309. 7x smaller. Open weights.
The trick: train on boundaries, not just semantics.
Robbyant releases LingBot-Vision, the world’s first spatially-native visual foundation model, a shift in the visual pre-training paradigm
Most vision foundation models are trained to become stable against visual changes, so they get very good at saying “cat,” “chair,” or “table” even when lighting or angle changes.
That is useful for recognition, but robotics needs something more annoying and physical: where the object starts, where it ends, where the thin cable is, where the transparent cup breaks depth sensing, and which boundary should stay stable as the robot moves.
LingBot-Vision is trained around boundaries. It learns boundary-bearing visual tokens by itself, without human labels, without an external edge detector, and without starting from another pretrained backbone.
Those boundary tokens then become training targets for dense visual features, so the model learns scene structure at the patch level rather than only learning high-level semantics.
The result is a frozen visual backbone that does well on depth, segmentation, and video object tracking, including 0.296 RMSE on NYUv2 depth, ahead of 7B-parameter DINOv3 at 0.309, while using far fewer parameters.
Similar Articles
@rohanpaul_ai: The robot’s “eyes” just received a big upgrade. LingBot-Depth 2.0, a depth-completion model with half the depth error j…
LingBot-Depth 2.0 is a depth-completion model that halves depth error, excels on transparent objects like glass and mirrors, and tops 12 out of 16 benchmarks, powered by the open-source LingBot-Vision backbone.
LingBot-Vision: masked boundary modeling for self-supervised pretraining (0.296 NYUv2 linear-probe RMSE at 1.1B vs 0.309 for DINOv3-7B, trails on ImageNet); weights in 4 sizes[R]
LingBot-Vision introduces masked boundary modeling for self-supervised pretraining, achieving a 0.296 RMSE on NYUv2 linear-probe with 1.1B parameters versus 0.309 for DINOv3-7B, though it trails on ImageNet; weights are released in four sizes.
@AdinaYakup: LingBot Vision A self-supervised vision backbone family for dense spatial perception from Ant Group @robbyant_brain - A…
LingBot Vision, a self-supervised vision backbone family from Ant Group, uses masked boundary modeling to achieve state-of-the-art performance on dense spatial perception tasks, beating the larger DINOv3 model on NYU-Depth v2.
Ant Group released LingBot-Vision: DINO-family vision backbones in 4 sizes, and the 0.3B ViT-L matches DINOv3-7B on NYUv2 depth with ~23x fewer params
Ant Group released LingBot-Vision, a family of DINO-style vision backbones in 4 sizes; the 0.3B ViT-L matches DINOv3-7B on NYUv2 depth with ~23x fewer parameters, showcasing significant efficiency gains.
Meta ships DINOv3 behind an access gate under its own license. Ant's Robbyant just shipped a full vision backbone family under Apache-2.0. What happens when perception goes free and small?
Robbyant, an embodied AI company under Ant Group, released LingBot-Vision, a self-supervised vision backbone family ranging from 21M to 1.1B parameters, under Apache-2.0. It matches or beats DINOv3 on several depth and segmentation benchmarks despite using less than one third of the training data, highlighting a push for open perception models.