# Image Classification on ImageNet (Accuracy and Performance) Source: [Image Classification on ImageNet (Accuracy and Performance)]() Collection index: [llms.txt]() Data updated: 2026-04-24T06:48:34.714Z Evaluation of Image Classification performance on the standard ImageNet dataset, reporting standard accuracy alongside inference speed (throughput, latency) and computational efficiency (GFLOPS). 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 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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. - [Throughput (Img/s)](): Higher is better; unit: SCALAR. - [GFLOPS](): Lower is better; unit: SIZE. - [Training Time (h)](): Lower is better; unit: TIME. - [Speedup (%)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | ViT-LMAE | 86.4 | Not recorded | 68fa7c501e209131cea8d818 | | APT | 86.3 | Not recorded | 68fa7c501e209131cea8d818 | | ViT-LMAE | 86.1 | Not recorded | 68fa7c501e209131cea8d818 | | APT | 86.1 | Not recorded | 68fa7c501e209131cea8d818 | | Resizing | 86 | Not recorded | 68fa7c501e209131cea8d818 | | Resizing | 85.9 | Not recorded | 68fa7c501e209131cea8d818 | | Random | 85.8 | Not recorded | 68fa7c501e209131cea8d818 | | Random | 85.5 | Not recorded | 68fa7c501e209131cea8d818 | | ViT-BMAE | 84.2 | Not recorded | 68fa7c501e209131cea8d818 | | APT | 84.2 | Not recorded | 68fa7c501e209131cea8d818 | | Resizing | 83.9 | Not recorded | 68fa7c501e209131cea8d818 | | Random | 83.4 | Not recorded | 68fa7c501e209131cea8d818 | ### ViT-LMAE Accuracy: 86.4. Result ID: 69eb11a947f34ce924d2a62a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"448/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### APT Accuracy: 86.3. Result ID: 69eb11a947f34ce924d2a62d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"448/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### ViT-LMAE Accuracy: 86.1. Result ID: 69eb11a947f34ce924d2a626. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"336/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### APT Accuracy: 86.1. Result ID: 69eb11a947f34ce924d2a629. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"336/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### Resizing Accuracy: 86. Result ID: 69eb11a947f34ce924d2a62c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"448/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### Resizing Accuracy: 85.9. Result ID: 69eb11a947f34ce924d2a628. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"336/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### Random Accuracy: 85.8. Result ID: 69eb11a947f34ce924d2a62b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Res/Patch":"448/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### Random Accuracy: 85.5. Result ID: 69eb11a947f34ce924d2a627. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Res/Patch":"336/14","Backbone":"ViT-L","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### ViT-BMAE Accuracy: 84.2. Result ID: 69eb11a947f34ce924d2a622. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"384/16","Backbone":"ViT-B","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### APT Accuracy: 84.2. Result ID: 69eb11a947f34ce924d2a625. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"384/16","Backbone":"ViT-B","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### Resizing Accuracy: 83.9. Result ID: 69eb11a947f34ce924d2a624. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Res/Patch":"384/16","Backbone":"ViT-B","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. ### Random Accuracy: 83.4. Result ID: 69eb11a947f34ce924d2a623. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Res/Patch":"384/16","Backbone":"ViT-B","Pre-training":"MAE","Fine-Tuning Protocol":"Full Fine-Tuning"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.