# Image Classification on MNIST (test) (Accuracy and Hardware Metrics)

Source: [Image Classification on MNIST (test) (Accuracy and Hardware Metrics)](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-accuracy-and-hardware-metrics>)

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

Data updated: 2026-03-26T14:37:46.479Z

Evaluation of image classification performance on the MNIST test set, reporting standard accuracy alongside hardware utilization (LUTs, FFs, BRAMs, DSP), latency, and power consumption.

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 | Classification |
| Dataset (including any stated split) | MNIST (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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-mnist-test-accuracy-and-hardware-metrics.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [LUTs](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-accuracy-and-hardware-metrics.md?metric=LUTs>): Lower is better; unit: SIZE.

- [FFs](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-accuracy-and-hardware-metrics.md?metric=FFs>): Lower is better; unit: SIZE.

- [BRAMs](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-accuracy-and-hardware-metrics.md?metric=BRAMs>): Lower is better; unit: SIZE.

- [DSPs](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-accuracy-and-hardware-metrics.md?metric=DSPs>): Lower is better; unit: SIZE.

- [Latency](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-accuracy-and-hardware-metrics.md?metric=Latency>): Lower is better; unit: TIME.

- [Total Power (W)](<https://www.sota2.com/research/sota/image-classification-on-mnist-test-accuracy-and-hardware-metrics.md?metric=Total%20Power%20(W)>): Lower is better; unit: OTHER.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DTM | 95.1 | Not recorded | 69c4bdb90636e5bf4afc779e |
| BTHOWeN | 93.4 | Not recorded | 69c4bdb90636e5bf4afc779e |
| TsetlinWiSARD-300 | 92.1 | Not recorded | 69c4bdb90636e5bf4afc779e |
| LogicWiSARD | 91 | Not recorded | 69c4bdb90636e5bf4afc779e |
| Hashed WNN | 90.73 | Not recorded | 69c4bdb90636e5bf4afc779e |
| TsetlinWiSARD-150 | 88.02 | Not recorded | 69c4bdb90636e5bf4afc779e |

### DTM

Accuracy: 95.1. Result ID: 69c544b347f34ce924af38a3.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

Recorded evaluation configuration: {"Platform":"XC7Z020","Algorithm":"TM","Purpose":"Training"}. This record does not assert independent reproduction.

### BTHOWeN

Accuracy: 93.4. Result ID: 69c544b347f34ce924af38a2.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf51e209131cea313c6.md>)

Recorded evaluation configuration: {"Platform":"XC7Z020","Algorithm":"WiSARD","Purpose":"Inference"}. This record does not assert independent reproduction.

### TsetlinWiSARD-300

Accuracy: 92.1. Result ID: 69c544b347f34ce924af389f.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

Recorded evaluation configuration: {"Platform":"XC7Z020","Algorithm":"WiSARD","Purpose":"Training"}. This record does not assert independent reproduction.

### LogicWiSARD

Accuracy: 91. Result ID: 69c544b347f34ce924af38a1.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

Recorded evaluation configuration: {"Platform":"XC7Z010","Algorithm":"WiSARD","Purpose":"Inference"}. This record does not assert independent reproduction.

### Hashed WNN

Accuracy: 90.73. Result ID: 69c544b347f34ce924af38a0.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

Recorded evaluation configuration: {"Platform":"XC7Z020","Algorithm":"WiSARD","Purpose":"Training"}. This record does not assert independent reproduction.

### TsetlinWiSARD-150

Accuracy: 88.02. Result ID: 69c544b347f34ce924af389e.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bdb90636e5bf4afc779e.md>)

Recorded evaluation configuration: {"Platform":"XC7Z020","Algorithm":"WiSARD","Purpose":"Training"}. This record does not assert independent reproduction.

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