| Benchmark | Model | Rank | Results |
|---|---|---|---|
| image-classification-on-cifar-100 | ViT-B-16 (ImageNet-21K-P pretrain) | #4 | Percentage correct: 94.2 |
| image-classification-on-stanford-cars | TResNet-L-V2 | #1 | Accuracy: 96.32 |
| multi-label-classification-on-ms-coco | TResNet-L-V2, (ImageNet-21K-P pretraining, resolution 640) | #10 | mAP: 89.8 |
| multi-label-classification-on-ms-coco | TResNet-L-V2, (ImageNet-21K-P pretraining, resolution 448) | #13 | mAP: 88.4 |
| multi-label-classification-on-pascal-voc-2007 | ViT-B-16 (ImageNet-21K pretrained) | #16 | mAP: 93.1 |