# Image Classification on CIFAR100 (test) (Top-K Accuracy)

Source: [Image Classification on CIFAR100 (test) (Top-K Accuracy)](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-top-k-accuracy>)

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

Data updated: 2026-02-26T06:06:40.766Z

Standard evaluation of Image Classification performance on the CIFAR-100 test set, reporting Top-1, Top-3, and Top-5 classification 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) | CIFAR100 (test) |
| Metric | Top-1 Acc |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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 Acc](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-top-k-accuracy.md?metric=Top-1%20Acc>): Higher is better; unit: PERCENT.

- [Top-3 Acc](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-top-k-accuracy.md?metric=Top-3%20Acc>): Higher is better; unit: PERCENT.

- [Top-5 Acc](<https://www.sota2.com/research/sota/image-classification-on-cifar100-test-top-k-accuracy.md?metric=Top-5%20Acc>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Top-1 Acc | Result date | Report paper |
| --- | --- | --- | --- |
| OURS\_GC | 22.7 | Not recorded | 68fa6ec31e209131cea41d3c |
| OURS\_Sep | 19.37 | Not recorded | 68fa6ec31e209131cea41d3c |
| CyCLIP | 15.09 | Not recorded | 68fa6ec31e209131cea41d3c |
| CLIP | 15.05 | Not recorded | 68fa6ec31e209131cea41d3c |
| OURS\_Br | 14.22 | Not recorded | 68fa6ec31e209131cea41d3c |

### OURS\_GC

Top-1 Acc: 22.7. Result ID: 699055e83a6c314849ffce7d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

Recorded evaluation configuration: {"Zero-shot":true,"Pre-training dataset":"CC3M","regularizer":"geometry consistency"}. This record does not assert independent reproduction.

### OURS\_Sep

Top-1 Acc: 19.37. Result ID: 699055e83a6c314849ffce7b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

Recorded evaluation configuration: {"Zero-shot":true,"Pre-training dataset":"CC3M","regularizer":"feature-separation"}. This record does not assert independent reproduction.

### CyCLIP

Top-1 Acc: 15.09. Result ID: 699055e83a6c314849ffce7a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dc51e209131cea3c046.md>)

[Implementation](<https://github.com/goel-shashank/CyCLIP>)

Recorded evaluation configuration: {"Zero-shot":true,"Pre-training dataset":"CC3M"}. This record does not assert independent reproduction.

### CLIP

Top-1 Acc: 15.05. Result ID: 699055e83a6c314849ffce79.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Zero-shot":true,"Pre-training dataset":"CC3M"}. This record does not assert independent reproduction.

### OURS\_Br

Top-1 Acc: 14.22. Result ID: 699055e83a6c314849ffce7c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d3c.md>)

Recorded evaluation configuration: {"Zero-shot":true,"Pre-training dataset":"CC3M","regularizer":"Brownian bridge"}. This record does not assert independent reproduction.

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