| Benchmark | Model | Rank | Results |
|---|---|---|---|
| self-supervised-image-classification-on-imagenet | DINOv2+reg (ViT-g/14) | #1 | Top 1 Accuracy: 87.1Number of Params: 1100M |
| linear-probe-semantic-segmentation-on-ade20k | DINOv2 with Registers (ViT-L/14) + Linear Probe | #8 | mIoU: 47.9 |
| image-classification-on-imagenet | DINOv2 with Registers (ViT-L/14) | #312 | Top 1 Accuracy: 84.8% |