# Image Classification on CIFAR-100 (Accuracy and Calibration)

Source: [Image Classification on CIFAR-100 (Accuracy and Calibration)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-accuracy-and-calibration>)

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

Data updated: 2026-06-11T11:08:57.507Z

Evaluation of standard Image Classification performance (Accuracy) on the CIFAR-100 dataset, including calibration metrics such as Confidence and Expected Calibration Error (ECE).

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) | CIFAR-100 |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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-cifar-100-accuracy-and-calibration.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Confidence](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-accuracy-and-calibration.md?metric=Confidence>): Higher is better; unit: PERCENT.

- [ECE](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-accuracy-and-calibration.md?metric=ECE>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ConvNeXt-Large | 92.17 | Not recorded | 6a2a414bd33dfb23e78394fa |
| ViT-B/16 | 91.51 | Not recorded | 6a2a414bd33dfb23e78394fa |
| MobileNetV2 (KD + MixUp) | 86.3 | Not recorded | 6a2a414bd33dfb23e78394fa |
| MobileNetV2 (KD + MixUp) | 85.8 | Not recorded | 6a2a414bd33dfb23e78394fa |
| MobileNetV2 (KD + MixUp) | 84.1 | Not recorded | 6a2a414bd33dfb23e78394fa |
| ResNet152V2 | 82.57 | Not recorded | 6a2a414bd33dfb23e78394fa |
| ConvNeXt-Tiny | 81.96 | Not recorded | 6a2a414bd33dfb23e78394fa |
| MobileNetV2 (KD + MixUp) | 81.1 | Not recorded | 6a2a414bd33dfb23e78394fa |
| MobileNetV2 Baseline | 81.05 | Not recorded | 6a2a414bd33dfb23e78394fa |

### ConvNeXt-Large

Accuracy: 92.17. Result ID: 6a2a972efbc373592c28bad7.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Role":"Teacher"}. This record does not assert independent reproduction.

### ViT-B/16

Accuracy: 91.51. Result ID: 6a2a972efbc373592c28bad8.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Role":"Teacher"}. This record does not assert independent reproduction.

### MobileNetV2 (KD + MixUp)

Accuracy: 86.3. Result ID: 6a2a972efbc373592c28badb.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Architecture":"MobileNetV2","Teacher Model":"ConvNeXt-Large","Training Protocol":"KD + MixUp"}. This record does not assert independent reproduction.

### MobileNetV2 (KD + MixUp)

Accuracy: 85.8. Result ID: 6a2a972efbc373592c28badc.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Architecture":"MobileNetV2","Teacher Model":"ViT-B/16","Training Protocol":"KD + MixUp"}. This record does not assert independent reproduction.

### MobileNetV2 (KD + MixUp)

Accuracy: 84.1. Result ID: 6a2a972efbc373592c28badd.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Architecture":"MobileNetV2","Teacher Model":"ResNet152V2","Training Protocol":"KD + MixUp"}. This record does not assert independent reproduction.

### ResNet152V2

Accuracy: 82.57. Result ID: 6a2a972efbc373592c28bad9.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Role":"Teacher"}. This record does not assert independent reproduction.

### ConvNeXt-Tiny

Accuracy: 81.96. Result ID: 6a2a972efbc373592c28bada.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Role":"Teacher"}. This record does not assert independent reproduction.

### MobileNetV2 (KD + MixUp)

Accuracy: 81.1. Result ID: 6a2a972efbc373592c28bade.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Architecture":"MobileNetV2","Teacher Model":"ConvNeXt-Tiny","Training Protocol":"KD + MixUp"}. This record does not assert independent reproduction.

### MobileNetV2 Baseline

Accuracy: 81.05. Result ID: 6a2a972efbc373592c28badf.

[Reporting paper](<https://www.sota2.com/research/paper/6a2a414bd33dfb23e78394fa.md>)

Recorded evaluation configuration: {"Architecture":"MobileNetV2","Training Protocol":"Baseline"}. This record does not assert independent reproduction.

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