# Image Classification on ImageNet-1K 22K pre-trained (val)

Source: [Image Classification on ImageNet-1K 22K pre-trained (val)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-22k-pre-trained-val>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:07:10.825Z

Evaluation of Image Classification performance on the validation split of ImageNet-1K, specifically for models pre-trained on the ImageNet-22K dataset.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Image Classification |
| Dataset (including any stated split) | ImageNet-1K 22K pre-trained (val) |
| Metric | Top-1 Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Top-1 Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-22k-pre-trained-val.md?metric=Top-1%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Top-1 Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Swin-L | 87.3 | Not recorded | 68fa6a8c1e209131cea28b62 |
| Swin-B | 86.4 | Not recorded | 68fa6a8c1e209131cea28b62 |
| R-152x4 | 85.4 | Not recorded | 68fa6a8c1e209131cea28b62 |
| ViT-L/16 | 85.2 | Not recorded | 68fa6a8c1e209131cea28b62 |
| Swin-B | 85.2 | Not recorded | 68fa6a8c1e209131cea28b62 |
| R-101x3 | 84.4 | Not recorded | 68fa6a8c1e209131cea28b62 |
| ViT-B/16 | 84 | Not recorded | 68fa6a8c1e209131cea28b62 |

### Swin-L

Top-1 Accuracy: 87.3. Result ID: 699ab8035698665ec445f3d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Implementation](<https://github.com/microsoft/Swin-Transformer>)

Recorded evaluation configuration: {"image size":"384x384","#param.":"197M","FLOPs":"103.9G","throughput (image/s)":"42.1"}. This record does not assert independent reproduction.

### Swin-B

Top-1 Accuracy: 86.4. Result ID: 699ab8035698665ec445f3d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Implementation](<https://github.com/microsoft/Swin-Transformer>)

Recorded evaluation configuration: {"image size":"384x384","#param.":"88M","FLOPs":"47.0G","throughput (image/s)":"84.7"}. This record does not assert independent reproduction.

### R-152x4

Top-1 Accuracy: 85.4. Result ID: 699ab8035698665ec445f3cc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa677a1e209131cea16a7e.md>)

[Implementation](<https://github.com/tensorflow/tpu/tree/master/models/official/retinanet>)

Recorded evaluation configuration: {"image size":"480x480","#param.":"937M","FLOPs":"840.5G","throughput (image/s)":null}. This record does not assert independent reproduction.

### ViT-L/16

Top-1 Accuracy: 85.2. Result ID: 699ab8035698665ec445f3ce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0b1e209131cea25b04.md>)

[Implementation](<https://github.com/google-research/vision_transformer>)

Recorded evaluation configuration: {"image size":"384x384","#param.":"307M","FLOPs":"190.7G","throughput (image/s)":"27.3"}. This record does not assert independent reproduction.

### Swin-B

Top-1 Accuracy: 85.2. Result ID: 699ab8035698665ec445f3cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Implementation](<https://github.com/microsoft/Swin-Transformer>)

Recorded evaluation configuration: {"image size":"224x224","#param.":"88M","FLOPs":"15.4G","throughput (image/s)":"278.1"}. This record does not assert independent reproduction.

### R-101x3

Top-1 Accuracy: 84.4. Result ID: 699ab8035698665ec445f3cb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa677a1e209131cea16a7e.md>)

[Implementation](<https://github.com/tensorflow/tpu/tree/master/models/official/retinanet>)

Recorded evaluation configuration: {"image size":"384x384","#param.":"388M","FLOPs":"204.6G","throughput (image/s)":null}. This record does not assert independent reproduction.

### ViT-B/16

Top-1 Accuracy: 84. Result ID: 699ab8035698665ec445f3cd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0b1e209131cea25b04.md>)

[Implementation](<https://github.com/google-research/vision_transformer>)

Recorded evaluation configuration: {"image size":"384x384","#param.":"86M","FLOPs":"55.4G","throughput (image/s)":"85.9"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.