| image-classification-on-imagenet | ConvNeXt V2-H (extra data) | #30 | Top 1 Accuracy: 88.9% |
| instance-segmentation-on-coco-minival | ConvNeXt V2-H + Mask R-CNN | #34 | mask AP: 48.9 |
| semantic-segmentation-on-ade20k | ConvNeXt V2-H (FCMAE) | #52 | Validation mIoU: 55 |
| semantic-segmentation-on-ade20k | Swin V2-H | #68 | Validation mIoU: 54.2 |
| semantic-segmentation-on-ade20k | ConvNeXt V2-L | #79 | Validation mIoU: 53.7 |
| semantic-segmentation-on-ade20k | Swin-L | #84 | Validation mIoU: 53.5 |
| semantic-segmentation-on-ade20k | Swin-B | #91 | Validation mIoU: 52.8 |
| semantic-segmentation-on-ade20k | ConvNeXt V2-B | #94 | Validation mIoU: 52.1 |
| semantic-segmentation-on-ade20k | ConvNeXt V2-L (Supervised) | #104 | Validation mIoU: 51.6 |
| semantic-segmentation-on-ade20k | ConvNeXt V1-L | #122 | Validation mIoU: 50.5 |
| semantic-segmentation-on-ade20k | ConvNeXt V1-B | #135 | Validation mIoU: 49.9 |
| object-detection-on-coco-val2017 | ConvNeXt V2-H + Mask R-CNN | #96 | box AP: 55.7 |