# Fine-grained Image Classification on Flowers-102 (test)

Source: [Fine-grained Image Classification on Flowers-102 (test)](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-flowers-102-test>)

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

Data updated: 2026-02-26T05:51:52.758Z

Evaluation of Fine-grained Image Classification performance on the test split of the Flowers-102 dataset, reporting Mean 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 | Fine-grained Image Classification |
| Dataset (including any stated split) | Flowers-102 (test) |
| Metric | Mean Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Mean Accuracy](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-flowers-102-test.md?metric=Mean%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Mean Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FastViT-SA12 | 59.9 | Not recorded | 698abb63c97a0b3e3b5dddee |
| TwistNet-18 | 58.5 | Not recorded | 698abb63c97a0b3e3b5dddee |
| RepViT-M1.5 | 51.6 | Not recorded | 698abb63c97a0b3e3b5dddee |
| Swin-Tiny | 48.8 | Not recorded | 698abb63c97a0b3e3b5dddee |
| ConvNeXtV2-Nano | 46.1 | Not recorded | 698abb63c97a0b3e3b5dddee |
| ResNet-18 | 43.6 | Not recorded | 698abb63c97a0b3e3b5dddee |
| SE-ResNet-18 | 40.5 | Not recorded | 698abb63c97a0b3e3b5dddee |
| ConvNeXt-Tiny | 7.5 | Not recorded | 698abb63c97a0b3e3b5dddee |

### FastViT-SA12

Mean Accuracy: 59.9. Result ID: 699730125698665ec47f342c.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70541e209131cea4a484.md>)

[Implementation](<https://github.com/apple/ml-fastvit>)

Recorded evaluation configuration: {"Params":"10.60M","FLOPs":"1.50G","Pre-training":"None"}. This record does not assert independent reproduction.

### TwistNet-18

Mean Accuracy: 58.5. Result ID: 699730125698665ec47f342e.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

Recorded evaluation configuration: {"Params":"11.59M","FLOPs":"1.85G","Pre-training":"None"}. This record does not assert independent reproduction.

### RepViT-M1.5

Mean Accuracy: 51.6. Result ID: 699730125698665ec47f342d.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72841e209131cea56f7a.md>)

[Implementation](<https://github.com/jameslahm/RepViT>)

Recorded evaluation configuration: {"Params":"13.67M","FLOPs":"2.31G","Pre-training":"None"}. This record does not assert independent reproduction.

### Swin-Tiny

Mean Accuracy: 48.8. Result ID: 699730125698665ec47f3430.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

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

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

Recorded evaluation configuration: {"Params":"27.56M","FLOPs":"4.51G","Pre-training":"None"}. This record does not assert independent reproduction.

### ConvNeXtV2-Nano

Mean Accuracy: 46.1. Result ID: 699730125698665ec47f342b.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e461e209131cea3ee37.md>)

[Implementation](<https://github.com/facebookresearch/ConvNeXt-V2>)

Recorded evaluation configuration: {"Params":"15.01M","FLOPs":"2.45G","Pre-training":"None"}. This record does not assert independent reproduction.

### ResNet-18

Mean Accuracy: 43.6. Result ID: 699730125698665ec47f3429.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

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

Recorded evaluation configuration: {"Params":"11.20M","FLOPs":"1.82G","Pre-training":"None"}. This record does not assert independent reproduction.

### SE-ResNet-18

Mean Accuracy: 40.5. Result ID: 699730125698665ec47f342a.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65bd1e209131cea0c0f9.md>)

[Implementation](<https://github.com/hujie-frank/SENet>)

Recorded evaluation configuration: {"Params":"11.29M","FLOPs":"1.82G","Pre-training":"None"}. This record does not assert independent reproduction.

### ConvNeXt-Tiny

Mean Accuracy: 7.5. Result ID: 699730125698665ec47f342f.

[Reporting paper](<https://www.sota2.com/research/paper/698abb63c97a0b3e3b5dddee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf21e209131cea31132.md>)

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

Recorded evaluation configuration: {"Params":"27.86M","FLOPs":"4.47G","Pre-training":"None"}. This record does not assert independent reproduction.

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