# Image Classification on ImageNet (test) (Accuracy and Efficiency)

Source: [Image Classification on ImageNet (test) (Accuracy and Efficiency)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-accuracy-and-efficiency>)

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

Data updated: 2026-02-26T05:57:16.922Z

Evaluation of Image Classification performance on the test split of ImageNet, reporting Top-1 Accuracy alongside model efficiency metrics (Parameter Count % and FLOPs %).

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 (test) |
| Metric | Params (%) |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 4 of 4 |
| 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

- [Params (%)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-accuracy-and-efficiency.md?metric=Params%20(%25)>): Lower is better; unit: PERCENT.

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

- [FLOPs (%)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-test-accuracy-and-efficiency.md?metric=FLOPs%20(%25)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Params (%) | Result date | Report paper |
| --- | --- | --- | --- |
| GAL-0.5 | 17.2 | Not recorded | 68fa69e51e209131cea24a5d |
| PFEC + CLR | 17.6 | Not recorded | 68fa69e51e209131cea24a5d |
| GAL-1 | 42.6 | Not recorded | 68fa69e51e209131cea24a5d |
| PFEC + CLR | 43.1 | Not recorded | 68fa69e51e209131cea24a5d |

### GAL-0.5

Params (%): 17.2. Result ID: 698def95ffc3901de099a64b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69e51e209131cea24a5d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65811e209131cea0a9eb.md>)

Recorded evaluation configuration: {"Model":"ResNet-50","Unpruned Top-1":76.15}. This record does not assert independent reproduction.

### PFEC + CLR

Params (%): 17.6. Result ID: 698def95ffc3901de099a64c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69e51e209131cea24a5d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69e51e209131cea24a5d.md>)

Recorded evaluation configuration: {"Model":"ResNet-50","Unpruned Top-1":76.15,"Target Complexity":"GAL-0.5 level"}. This record does not assert independent reproduction.

### GAL-1

Params (%): 42.6. Result ID: 698def95ffc3901de099a64d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69e51e209131cea24a5d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65811e209131cea0a9eb.md>)

Recorded evaluation configuration: {"Model":"ResNet-50","Unpruned Top-1":76.15}. This record does not assert independent reproduction.

### PFEC + CLR

Params (%): 43.1. Result ID: 698def95ffc3901de099a64e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69e51e209131cea24a5d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69e51e209131cea24a5d.md>)

Recorded evaluation configuration: {"Model":"ResNet-50","Unpruned Top-1":76.15,"Target Complexity":"GAL-1 level"}. This record does not assert independent reproduction.

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