# Image Classification on ImageNet-1K (Accuracy (P), Compression, Time, I' Score)

Source: [Image Classification on ImageNet-1K (Accuracy (P), Compression, Time, I' Score)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-accuracy-p-compression-time-i-score>)

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

Data updated: 2026-05-13T07:03:25.588Z

Evaluation of Image Classification performance on the ImageNet-1K dataset, reporting standard accuracy (P), compression ratio (C), inference time (T), and I' Score.

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-1K |
| Metric | Accuracy (P) |
| Unit | PERCENT |
| Direction | Higher 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

- [Accuracy (P)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-accuracy-p-compression-time-i-score.md?metric=Accuracy%20(P)>): Higher is better; unit: PERCENT.

- [Compression Ratio (C)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-accuracy-p-compression-time-i-score.md?metric=Compression%20Ratio%20(C)>): Higher is better; unit: SCALAR.

- [Inference Time (T) (ms)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-accuracy-p-compression-time-i-score.md?metric=Inference%20Time%20(T)%20(ms)>): Lower is better; unit: TIME.

- [I' Score](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-accuracy-p-compression-time-i-score.md?metric=I%27%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy (P) | Result date | Report paper |
| --- | --- | --- | --- |
| Uniform Baseline | 69.76 | Not recorded | 6a04058cdbc836b87d8274eb |
| HAQ | 69.7 | Not recorded | 6a04058cdbc836b87d8274eb |
| Uniform PTQ (8-bit) | 69.36 | Not recorded | 6a04058cdbc836b87d8274eb |
| QuIDE GA | 69.36 | Not recorded | 6a04058cdbc836b87d8274eb |

### Uniform Baseline

Accuracy (P): 69.76. Result ID: 6a0420b947f34ce924e97b32.

[Reporting paper](<https://www.sota2.com/research/paper/6a04058cdbc836b87d8274eb.md>)

Recorded evaluation configuration: {"Bit-width Topology":"32-32-32-32-32-32","Backbone":"ResNet-18"}. This record does not assert independent reproduction.

### HAQ

Accuracy (P): 69.7. Result ID: 6a0420b947f34ce924e97b34.

[Reporting paper](<https://www.sota2.com/research/paper/6a04058cdbc836b87d8274eb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65a41e209131cea0b895.md>)

Recorded evaluation configuration: {"Bit-width Topology":"16-16-16-8-16-16","Backbone":"ResNet-18","Note":"Re-implementation"}. This record does not assert independent reproduction.

### Uniform PTQ (8-bit)

Accuracy (P): 69.36. Result ID: 6a0420b947f34ce924e97b33.

[Reporting paper](<https://www.sota2.com/research/paper/6a04058cdbc836b87d8274eb.md>)

Recorded evaluation configuration: {"Bit-width Topology":"8-8-8-8-8-8","Backbone":"ResNet-18"}. This record does not assert independent reproduction.

### QuIDE GA

Accuracy (P): 69.36. Result ID: 6a0420b947f34ce924e97b35.

[Reporting paper](<https://www.sota2.com/research/paper/6a04058cdbc836b87d8274eb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a04058cdbc836b87d8274eb.md>)

Recorded evaluation configuration: {"Bit-width Topology":"8-8-8-8-8-16","Backbone":"ResNet-18"}. This record does not assert independent reproduction.

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