# Image Classification on CIFAR-10 (Accuracy, Norm. Energy, Runtime)

Source: [Image Classification on CIFAR-10 (Accuracy, Norm. Energy, Runtime)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-accuracy-norm-energy-runtime>)

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

Data updated: 2026-05-12T07:21:09.543Z

Evaluation of Image Classification performance on the CIFAR-10 dataset, reporting standard accuracy alongside energy consumption and inference runtime metrics.

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 |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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-accuracy-norm-energy-runtime.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Normalized Energy](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-accuracy-norm-energy-runtime.md?metric=Normalized%20Energy>): Lower is better; unit: PERCENT.

- [Runtime (/hour)](<https://www.sota2.com/research/sota/image-classification-on-cifar-10-accuracy-norm-energy-runtime.md?metric=Runtime%20(%2Fhour)>): Lower is better; unit: TIME.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| TRAM | 93.71 | Not recorded | 6a02b3eedbc836b87d82453d |
| FP32 | 93.65 | Not recorded | 6a02b3eedbc836b87d82453d |
| TRAM | 93.65 | Not recorded | 6a02b3eedbc836b87d82453d |
| AccMul | 93.56 | Not recorded | 6a02b3eedbc836b87d82453d |
| TRAM | 92.97 | Not recorded | 6a02b3eedbc836b87d82453d |
| TRAM | 92.79 | Not recorded | 6a02b3eedbc836b87d82453d |
| MARLIN-1 | 92.14 | Not recorded | 6a02b3eedbc836b87d82453d |
| MARLIN-2 | 91.7 | Not recorded | 6a02b3eedbc836b87d82453d |
| ALWANN-1 | 89.08 | Not recorded | 6a02b3eedbc836b87d82453d |
| ALWANN-2 | 88.58 | Not recorded | 6a02b3eedbc836b87d82453d |

### TRAM

Accuracy: 93.71. Result ID: 6a02d4db47f34ce924e53a8d.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8","Lambda (λ)":"1"}. This record does not assert independent reproduction.

### FP32

Accuracy: 93.65. Result ID: 6a02d4db47f34ce924e53a8b.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50"}. This record does not assert independent reproduction.

### TRAM

Accuracy: 93.65. Result ID: 6a02d4db47f34ce924e53a8e.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8","Lambda (λ)":"10"}. This record does not assert independent reproduction.

### AccMul

Accuracy: 93.56. Result ID: 6a02d4db47f34ce924e53a8c.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50"}. This record does not assert independent reproduction.

### TRAM

Accuracy: 92.97. Result ID: 6a02d4db47f34ce924e53a90.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8","Lambda (λ)":"1000"}. This record does not assert independent reproduction.

### TRAM

Accuracy: 92.79. Result ID: 6a02d4db47f34ce924e53a8f.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8","Lambda (λ)":"100"}. This record does not assert independent reproduction.

### MARLIN-1

Accuracy: 92.14. Result ID: 6a02d4db47f34ce924e53a91.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8"}. This record does not assert independent reproduction.

### MARLIN-2

Accuracy: 91.7. Result ID: 6a02d4db47f34ce924e53a92.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8"}. This record does not assert independent reproduction.

### ALWANN-1

Accuracy: 89.08. Result ID: 6a02d4db47f34ce924e53a93.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67011e209131cea13b09.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8"}. This record does not assert independent reproduction.

### ALWANN-2

Accuracy: 88.58. Result ID: 6a02d4db47f34ce924e53a94.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3eedbc836b87d82453d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67011e209131cea13b09.md>)

Recorded evaluation configuration: {"Model":"ResNet50","Quantization":"w8a8"}. This record does not assert independent reproduction.

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