# Image Classification on ImageNet ReaL 6 (val)

Source: [Image Classification on ImageNet ReaL 6 (val)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-real-6-val>)

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

Data updated: 2026-02-26T05:29:24.259Z

Evaluation of Image Classification performance on the validation split of the ImageNet ReaL 6 dataset, typically measured by Accuracy.

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 ReaL 6 (val) |
| Metric | Accuracy |
| Unit | SCALAR |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-real-6-val.md?metric=Accuracy>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SoViT-400m/14 | 0.91 | Not recorded | 68fa71d01e209131cea531ae |
| SoViT-400m/14 | 0.909 | Not recorded | 68fa71d01e209131cea531ae |
| ViT-g/14 | 0.909 | Not recorded | 68fa71d01e209131cea531ae |
| MAE-WSP | 0.909 | Not recorded | 68fa71d01e209131cea531ae |
| ViT-G/14 | 0.908 | Not recorded | 68fa71d01e209131cea531ae |
| ViT-L/16 | 0.904 | Not recorded | 68fa71d01e209131cea531ae |
| SoViT-400m/14 | 0.903 | Not recorded | 68fa71d01e209131cea531ae |

### SoViT-400m/14

Accuracy: 0.91. Result ID: 6991fd1e3a6c3148490be0a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

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

Recorded evaluation configuration: {"Pretraining":"JFT-3B","Input resolution":"518x518","Parameters":"428M","FLOPs":"1374G","Fine-tuning":true}. This record does not assert independent reproduction.

### SoViT-400m/14

Accuracy: 0.909. Result ID: 6991fd1e3a6c3148490be0a0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

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

Recorded evaluation configuration: {"Pretraining":"JFT-3B","Input resolution":"384x384","Parameters":"428M","FLOPs":"672G","Fine-tuning":true}. This record does not assert independent reproduction.

### ViT-g/14

Accuracy: 0.909. Result ID: 6991fd1e3a6c3148490be0a1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

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

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

Recorded evaluation configuration: {"Pretraining":"JFT-3B","Input resolution":"518x518","Parameters":"1011M","FLOPs":"3208G","Fine-tuning":true}. This record does not assert independent reproduction.

### MAE-WSP

Accuracy: 0.909. Result ID: 6991fd1e3a6c3148490be0a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71ee1e209131cea539db.md>)

[Implementation](<https://github.com/facebookresearch/maws>)

Recorded evaluation configuration: {"Pretraining":"IG-3B","Input resolution":"518x518","Parameters":"1890M","FLOPs":"5679G","Fine-tuning":true}. This record does not assert independent reproduction.

### ViT-G/14

Accuracy: 0.908. Result ID: 6991fd1e3a6c3148490be0a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cd91e209131cea363c2.md>)

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

Recorded evaluation configuration: {"Pretraining":"JFT-3B","Input resolution":"518x518","Parameters":"1882M","FLOPs":"5668G","Fine-tuning":true}. This record does not assert independent reproduction.

### ViT-L/16

Accuracy: 0.904. Result ID: 6991fd1e3a6c3148490be09f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

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

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

Recorded evaluation configuration: {"Pretraining":"JFT-3B","Input resolution":"384x384","Parameters":"303M","FLOPs":"383G","Fine-tuning":true}. This record does not assert independent reproduction.

### SoViT-400m/14

Accuracy: 0.903. Result ID: 6991fd1e3a6c3148490be09e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71d01e209131cea531ae.md>)

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

Recorded evaluation configuration: {"Pretraining":"JFT-3B","Input resolution":"224x224","Parameters":"428M","FLOPs":"221G","Fine-tuning":true}. This record does not assert independent reproduction.

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