# Image Classification on CIFAR-10 (test) (Accuracy, Overhead, and Speedup)

Source: [Image Classification on CIFAR-10 (test) (Accuracy, Overhead, and Speedup)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-overhead-and-speedup>)

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

Data updated: 2026-03-06T08:28:43.026Z

Evaluation of standard Image Classification accuracy on the CIFAR-10 test set, including associated computational efficiency metrics such as GPU hour overhead and model speedup.

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-10 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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-10-test-accuracy-overhead-and-speedup.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Overhead (GPU h)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-overhead-and-speedup.md?metric=Overhead%20(GPU%20h)>): Lower is better; unit: TIME.

- [Speedup](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-test-accuracy-overhead-and-speedup.md?metric=Speedup>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Full | 96.1 | Not recorded | 69aa5f722d603e6de3e41831 |
| Ours | 95.7 | Not recorded | 69aa5f722d603e6de3e41831 |

### Full

Accuracy: 96.1. Result ID: 69aa8fd96dbb7075b9544dd6.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f722d603e6de3e41831.md>)

Recorded evaluation configuration: {"Method":"Full"}. This record does not assert independent reproduction.

### Ours

Accuracy: 95.7. Result ID: 69aa8fd96dbb7075b9544dd7.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f722d603e6de3e41831.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69aa5f722d603e6de3e41831.md>)

Recorded evaluation configuration: {"Method":"Ours"}. This record does not assert independent reproduction.

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