# Image Classification on MNIST (Accuracy and Hardware Efficiency Metrics) Source: [Image Classification on MNIST (Accuracy and Hardware Efficiency Metrics)]() Collection index: [llms.txt]() Data updated: 2026-04-23T08:27:21.473Z Evaluation of Image Classification performance on the MNIST dataset, reporting standard Accuracy alongside detailed hardware efficiency metrics including inference latency and energy consumption, both absolute and equivalent 16nm scaled values. 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) | MNIST | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 7 of 7 | | 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](): Higher is better; unit: PERCENT. - [Inference Latency (s)](): Lower is better; unit: TIME. - [Equivalent 16nm Latency (ns)](): Lower is better; unit: TIME. - [Energy per Inference](): Lower is better; unit: OTHER. - [Equivalent 16nm Energy per Inference (pJ)](): Lower is better; unit: OTHER. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Datasheet | 99.6 | Not recorded | 69e85bec7d0172140ec811cd | | NeurIPS’24 | 99.23 | Not recorded | 69e85bec7d0172140ec811cd | | JSSC’19 | 97.9 | Not recorded | 69e85bec7d0172140ec811cd | | This Work | 97.49 | Not recorded | 69e85bec7d0172140ec811cd | | arXiv’25 | 97.42 | Not recorded | 69e85bec7d0172140ec811cd | | CICC’21 | 97.1 | Not recorded | 69e85bec7d0172140ec811cd | | FPGA’17 | 95.8 | Not recorded | 69e85bec7d0172140ec811cd | ### Datasheet Accuracy: 99.6. Result ID: 69e9d77f47f34ce924d0f439. [Reporting paper]() Recorded evaluation configuration: {"Implementation Technology":"40 nm CMOS"}. This record does not assert independent reproduction. ### NeurIPS’24 Accuracy: 99.23. Result ID: 69e9d77f47f34ce924d0f436. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Implementation Technology":"FPGA (XC7Z045)"}. This record does not assert independent reproduction. ### JSSC’19 Accuracy: 97.9. Result ID: 69e9d77f47f34ce924d0f43b. [Reporting paper]() Recorded evaluation configuration: {"Implementation Technology":"28 nm CMOS"}. This record does not assert independent reproduction. ### This Work Accuracy: 97.49. Result ID: 69e9d77f47f34ce924d0f435. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Implementation Technology":"130 nm CMOS"}. This record does not assert independent reproduction. ### arXiv’25 Accuracy: 97.42. Result ID: 69e9d77f47f34ce924d0f437. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Implementation Technology":"65 nm CMOS"}. This record does not assert independent reproduction. ### CICC’21 Accuracy: 97.1. Result ID: 69e9d77f47f34ce924d0f43a. [Reporting paper]() Recorded evaluation configuration: {"Implementation Technology":"28 nm CMOS"}. This record does not assert independent reproduction. ### FPGA’17 Accuracy: 95.8. Result ID: 69e9d77f47f34ce924d0f438. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Implementation Technology":"FPGA (Zynq ZC706)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.