ImageNet-R Dataset
| family | model | params (m) | pretrain | finetune | GFLOPs | Top-1 |
|---|---|---|---|---|---|---|
| Swin | Swin-T | 29.0 | IN-1k : Sup. : 300 | — : — : — | 4.5 | 41.3 |
| CoCA ViT | CoCA ViT-11M | 11.4 | IN-1k : Sup. : 300 | — : — : — | 2.2 | 47.8 |
| CAFormer | CAFormer-S18 | 26.0 | IN-1k : Sup. : 300 | — : — : — | 4.1 | 48.7 |
| CAFormer | CAFormer-S18 | 26.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 13.4 | 51.3 |
| FAN | FAN-L-Hybrid | 76.8 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 16.9 | 64.3 |
| TransNeXt | TransNeXt-S | 49.7 | IN-1k : Sup. : 300 | — : — : — | 10.3 | 52.5 |
| TransNeXt | TransNeXt-S | 49.7 | IN-1k : Sup. : 300 | IN-1k : 5 : 384 | 32.1 | 56.4 |
| CAFormer | CAFormer-B36 | 99.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 23.2 | 68.3 |
| CAFormer | CAFormer-B36 | 99.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 72.2 | 70.4 |
| FAN | FAN-L-Hybrid | 76.8 | IN-1k : Sup. + TL : 350 | — : — : — | 17.0 | 56.6 |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-1k : Sup. : 300 | — : — : — | 22.6 | 51.1 |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 66.5 | 52.2 |
| ConvNeXt | ConvNeXt-L | 198.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 101.0 | 66.7 |
| FAN | FAN-S-Hybrid | 26.0 | IN-1k : Sup. + TL : 350 | — : — : — | 6.8 | 51.8 |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 22.6 | 65.3 |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 66.5 | 66.5 |
| CAFormer | CAFormer-M36 | 56.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 13.2 | 63.4 |
| CAFormer | CAFormer-M36 | 56.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 42.0 | 65.3 |
| FAN | FAN-S-ViT | 28.0 | IN-1k : Sup. : 300 | — : — : — | 5.3 | 50.4 |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 7.6 | 58.9 |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 22.4 | 59.9 |
| ConvNeXt | ConvNeXt-B | 89.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 45.1 | 64.9 |
| CAFormer | CAFormer-S36 | 39.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 8.0 | 60.7 |
| CAFormer | CAFormer-S36 | 39.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 26.0 | 63.0 |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-1k : Sup. : 300 | — : — : — | 12.8 | 51.0 |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 37.7 | 52.2 |
| ConvNeXt | ConvNeXt-T | 29.0 | IN-1k : Sup. : 300 | — : — : — | 4.5 | 47.2 |
| FAN | FAN-B-ViT | 54.0 | IN-1k : Sup. : 300 | — : — : — | 10.4 | 51.8 |
| CAFormer | CAFormer-B36 | 99.0 | IN-1k : Sup. : 300 | — : — : — | 23.2 | 53.9 |
| CAFormer | CAFormer-B36 | 99.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 72.2 | 55.0 |
| FAN | FAN-B-Hybrid | 50.0 | IN-1k : Sup. : 300 | — : — : — | 11.3 | 52.9 |
| CAFormer | CAFormer-M36 | 56.0 | IN-1k : Sup. : 300 | — : — : — | 13.2 | 51.7 |
| CAFormer | CAFormer-M36 | 56.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 42.0 | 55.0 |
| ConvNeXt | ConvNeXt-XL | 350.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 179.0 | 68.2 |
| TransNeXt | TransNeXt-T | 28.2 | IN-1k : Sup. : 300 | — : — : — | 5.7 | 49.6 |
| SSViT | SSViT-T | 15.0 | IN-1k : Sup. : 300 | — : — : — | 2.4 | 45.6 |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 3.9 | 53.4 |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 11.6 | 55.0 |
| CoCA ViT | CoCA ViT-28M | 27.8 | IN-1k : Sup. : 300 | — : — : — | 4.9 | 51.1 |
| TransNeXt | TransNeXt-B | 89.7 | IN-1k : Sup. : 300 | — : — : — | 18.4 | 53.9 |
| TransNeXt | TransNeXt-B | 89.7 | IN-1k : Sup. : 300 | IN-1k : 5 : 384 | 56.3 | 57.7 |
| CAFormer | CAFormer-S36 | 39.0 | IN-1k : Sup. : 300 | — : — : — | 8.0 | 51.7 |
| CAFormer | CAFormer-S36 | 39.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 26.0 | 54.5 |
| SSViT | SSViT-S | 27.0 | IN-1k : Sup. : 300 | — : — : — | 4.4 | 51.0 |
| ConvNeXt | ConvNeXt-B | 89.0 | IN-1k : Sup. : 300 | — : — : — | 15.4 | 51.3 |
| SSViT | SSViT-B | 57.0 | IN-1k : Sup. : 300 | — : — : — | 9.6 | 55.6 |
| FAN | FAN-B-Hybrid | 50.0 | IN-1k : Sup. + TL : 350 | — : — : — | 11.4 | 55.3 |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 12.8 | 60.9 |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 37.7 | 61.8 |
| SSViT | SSViT-L | 100.0 | IN-1k : Sup. : 300 | — : — : — | 18.2 | 59.2 |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-1k : Sup. : 300 | — : — : — | 3.9 | 48.7 |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 11.6 | 50.7 |
| FAN | FAN-L-ViT | 81.0 | IN-1k : Sup. : 300 | — : — : — | 15.8 | 53.1 |
| FAN | FAN-L-Hybrid | 76.8 | IN-1k : Sup. : 300 | — : — : — | 16.9 | 53.2 |
| CAFormer | CAFormer-S18 | 26.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 4.1 | 54.1 |
| CAFormer | CAFormer-S18 | 26.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 13.4 | 55.9 |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-1k : Sup. : 300 | — : — : — | 7.6 | 50.8 |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 22.4 | 51.9 |
| CoCA ViT | CoCA ViT-21M | 20.6 | IN-1k : Sup. : 300 | — : — : — | 4.1 | 50.1 |
| Swin | Swin-B | 88.0 | IN-1k : Sup. : 300 | — : — : — | 15.4 | 46.6 |
| FAN | FAN-S-Hybrid | 26.0 | IN-1k : Sup. : 300 | — : — : — | 6.7 | 50.7 |
| TransNeXt | TransNeXt-Micro | 12.8 | IN-1k : Sup. : 300 | — : — : — | 2.7 | 45.8 |
| family | model | params (m) | pretrain | finetune | gflops | top-1 | top-5 |
|---|---|---|---|---|---|---|---|
| TransNeXt | TransNeXt-Micro | 12.8 | IN-1k : Sup. : 300 | — : — : — | 2.7 | 45.8 | — |
| TransNeXt | TransNeXt-T | 28.2 | IN-1k : Sup. : 300 | — : — : — | 5.7 | 49.6 | — |
| TransNeXt | TransNeXt-S | 49.7 | IN-1k : Sup. : 300 | — : — : — | 10.3 | 52.5 | — |
| TransNeXt | TransNeXt-S | 49.7 | IN-1k : Sup. : 300 | IN-1k : 5 : 384 | 32.1 | 56.4 | — |
| TransNeXt | TransNeXt-B | 89.7 | IN-1k : Sup. : 300 | — : — : — | 18.4 | 53.9 | — |
| TransNeXt | TransNeXt-B | 89.7 | IN-1k : Sup. : 300 | IN-1k : 5 : 384 | 56.3 | 57.7 | — |
| ConvNeXt | ConvNeXt-L | 198.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 101.0 | 66.7 | — |
| ConvNeXt | ConvNeXt-B | 89.0 | IN-1k : Sup. : 300 | — : — : — | 15.4 | 51.3 | — |
| ConvNeXt | ConvNeXt-B | 89.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 45.1 | 64.9 | — |
| ConvNeXt | ConvNeXt-T | 29.0 | IN-1k : Sup. : 300 | — : — : — | 4.5 | 47.2 | — |
| ConvNeXt | ConvNeXt-XL | 350.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 179.0 | 68.2 | — |
| Swin | Swin-T | 29.0 | IN-1k : Sup. : 300 | — : — : — | 4.5 | 41.3 | — |
| Swin | Swin-B | 88.0 | IN-1k : Sup. : 300 | — : — : — | 15.4 | 46.6 | — |
| CAFormer | CAFormer-S18 | 26.0 | IN-1k : Sup. : 300 | — : — : — | 4.1 | 48.7 | — |
| CAFormer | CAFormer-S18 | 26.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 13.4 | 51.3 | — |
| CAFormer | CAFormer-S18 | 26.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 4.1 | 54.1 | — |
| CAFormer | CAFormer-S18 | 26.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 13.4 | 55.9 | — |
| CAFormer | CAFormer-S36 | 39.0 | IN-1k : Sup. : 300 | — : — : — | 8.0 | 51.7 | — |
| CAFormer | CAFormer-S36 | 39.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 26.0 | 54.5 | — |
| CAFormer | CAFormer-S36 | 39.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 8.0 | 60.7 | — |
| CAFormer | CAFormer-S36 | 39.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 26.0 | 63.0 | — |
| CAFormer | CAFormer-M36 | 56.0 | IN-1k : Sup. : 300 | — : — : — | 13.2 | 51.7 | — |
| CAFormer | CAFormer-M36 | 56.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 42.0 | 55.0 | — |
| CAFormer | CAFormer-M36 | 56.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 13.2 | 63.4 | — |
| CAFormer | CAFormer-M36 | 56.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 42.0 | 65.3 | — |
| CAFormer | CAFormer-B36 | 99.0 | IN-1k : Sup. : 300 | — : — : — | 23.2 | 53.9 | — |
| CAFormer | CAFormer-B36 | 99.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 72.2 | 55.0 | — |
| CAFormer | CAFormer-B36 | 99.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 23.2 | 68.3 | — |
| CAFormer | CAFormer-B36 | 99.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 72.2 | 70.4 | — |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-1k : Sup. : 300 | — : — : — | 3.9 | 48.7 | — |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 11.6 | 50.7 | — |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 3.9 | 53.4 | — |
| ConvFormer | ConvFormer-S18 | 27.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 11.6 | 55.0 | — |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-1k : Sup. : 300 | — : — : — | 7.6 | 50.8 | — |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 22.4 | 51.9 | — |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 7.6 | 58.9 | — |
| ConvFormer | ConvFormer-S36 | 40.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 22.4 | 59.9 | — |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-1k : Sup. : 300 | — : — : — | 12.8 | 51.0 | — |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 37.7 | 52.2 | — |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 12.8 | 60.9 | — |
| ConvFormer | ConvFormer-M36 | 57.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 37.7 | 61.8 | — |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-1k : Sup. : 300 | — : — : — | 22.6 | 51.1 | — |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-1k : Sup. : 300 | IN-1k : 30 : 384 | 66.5 | 52.2 | — |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 22.6 | 65.3 | — |
| ConvFormer | ConvFormer-B36 | 100.0 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | 66.5 | 66.5 | — |
| FAN | FAN-S-ViT | 28.0 | IN-1k : Sup. : 300 | — : — : — | 5.3 | 50.4 | — |
| FAN | FAN-S-Hybrid | 26.0 | IN-1k : Sup. : 300 | — : — : — | 6.7 | 50.7 | — |
| FAN | FAN-S-Hybrid | 26.0 | IN-1k : Sup. + TL : 350 | — : — : — | 6.8 | 51.8 | — |
| FAN | FAN-B-ViT | 54.0 | IN-1k : Sup. : 300 | — : — : — | 10.4 | 51.8 | — |
| FAN | FAN-B-Hybrid | 50.0 | IN-1k : Sup. : 300 | — : — : — | 11.3 | 52.9 | — |
| FAN | FAN-B-Hybrid | 50.0 | IN-1k : Sup. + TL : 350 | — : — : — | 11.4 | 55.3 | — |
| FAN | FAN-L-ViT | 81.0 | IN-1k : Sup. : 300 | — : — : — | 15.8 | 53.1 | — |
| FAN | FAN-L-Hybrid | 76.8 | IN-1k : Sup. : 300 | — : — : — | 16.9 | 53.2 | — |
| FAN | FAN-L-Hybrid | 76.8 | IN-22k : Sup. : 90 | IN-1k : 30 : 224 | 16.9 | 64.3 | — |
| FAN | FAN-L-Hybrid | 76.8 | IN-22k : Sup. : 90 | IN-1k : 30 : 384 | None | 71.1 | — |
| FAN | FAN-L-Hybrid | 76.8 | IN-1k : Sup. + TL : 350 | — : — : — | 17.0 | 56.6 | — |
| SSViT | SSViT-T | 15.0 | IN-1k : Sup. : 300 | — : — : — | 2.4 | 45.6 | — |
| SSViT | SSViT-S | 27.0 | IN-1k : Sup. : 300 | — : — : — | 4.4 | 51.0 | — |
| SSViT | SSViT-B | 57.0 | IN-1k : Sup. : 300 | — : — : — | 9.6 | 55.6 | — |
| SSViT | SSViT-L | 100.0 | IN-1k : Sup. : 300 | — : — : — | 18.2 | 59.2 | — |
| CoCA ViT | CoCA ViT-11M | 11.4 | IN-1k : Sup. : 300 | — : — : — | 2.2 | 47.8 | — |
| CoCA ViT | CoCA ViT-21M | 20.6 | IN-1k : Sup. : 300 | — : — : — | 4.1 | 50.1 | — |
| CoCA ViT | CoCA ViT-28M | 27.8 | IN-1k : Sup. : 300 | — : — : — | 4.9 | 51.1 | — |