# Image Classification on ImageNet-1K (test) (Standard & Perturbation Accuracy)

Source: [Image Classification on ImageNet-1K (test) (Standard & Perturbation Accuracy)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-test-standard-and-perturbation-accuracy>)

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

Data updated: 2026-02-26T05:32:17.604Z

Evaluation of Image Classification performance on the test split of the ImageNet-1K dataset, reporting both standard accuracy and accuracy measured after applying perturbations.

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 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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-1k-test-standard-and-perturbation-accuracy.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Accuracy (Perturbed)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-test-standard-and-perturbation-accuracy.md?metric=Accuracy%20(Perturbed)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ViT-B/16 | 81.1 | Not recorded | 69705db4e25ad2fa9d1ed3d2 |
| ResNet-50 | 76.1 | Not recorded | 69705db4e25ad2fa9d1ed3d2 |
| GoogLeNet | 69.8 | Not recorded | 69705db4e25ad2fa9d1ed3d2 |

### ViT-B/16

Accuracy: 81.1. Result ID: 69992fed5698665ec49895c5.

[Reporting paper](<https://www.sota2.com/research/paper/69705db4e25ad2fa9d1ed3d2.md>)

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

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

Recorded evaluation configuration: {"Network":"ViT-B/16","# parameters":"86567656"}. This record does not assert independent reproduction.

### ResNet-50

Accuracy: 76.1. Result ID: 69992fed5698665ec49895c4.

[Reporting paper](<https://www.sota2.com/research/paper/69705db4e25ad2fa9d1ed3d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Network":"ResNet-50","# parameters":"25557032"}. This record does not assert independent reproduction.

### GoogLeNet

Accuracy: 69.8. Result ID: 69992fed5698665ec49895c3.

[Reporting paper](<https://www.sota2.com/research/paper/69705db4e25ad2fa9d1ed3d2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61801e209131ce9f4ab4.md>)

Recorded evaluation configuration: {"Network":"GoogLeNet","# parameters":"6624904"}. This record does not assert independent reproduction.

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