# Image Classification on ImageNet 1k V2 (test) Source: [Image Classification on ImageNet 1k V2 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:28:52.344Z Evaluation of Image Classification performance on the test split of the ImageNet 1k Version 2 dataset, typically measured by accuracy. 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 V2 (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | 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. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Lion | 75.84 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 75.1 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 74.06 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 73.81 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 69.19 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 68.14 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 67.68 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 66.73 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 64.81 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 64.24 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 64.14 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 64.02 | Not recorded | 68fa6f421e209131cea448e3 | | SGD | 63.93 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 63.09 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 61.87 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 55.94 | Not recorded | 68fa6f421e209131cea448e3 | | Lion | 55.75 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 55.01 | Not recorded | 68fa6f421e209131cea448e3 | | AdamW | 53.37 | Not recorded | 68fa6f421e209131cea448e3 | ### Lion Accuracy: 75.84. Result ID: 6990b0ff3a6c314849027add. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"ViT-L/16","Resolution":384,"Optimizer":"Lion","RandAug + Mixup":false,"Training Protocol":"Pre-train on ImageNet-21K then fine-tune on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 75.1. Result ID: 6990b0ff3a6c314849027adc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"ViT-L/16","Resolution":384,"Optimizer":"AdamW","RandAug + Mixup":false,"Training Protocol":"Pre-train on ImageNet-21K then fine-tune on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 74.06. Result ID: 6990b0ff3a6c314849027adb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"ViT-B/16","Resolution":384,"Optimizer":"Lion","RandAug + Mixup":false,"Training Protocol":"Pre-train on ImageNet-21K then fine-tune on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 73.81. Result ID: 6990b0ff3a6c314849027ada. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"ViT-B/16","Resolution":384,"Optimizer":"AdamW","RandAug + Mixup":false,"Training Protocol":"Pre-train on ImageNet-21K then fine-tune on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 69.19. Result ID: 6990b0ff3a6c314849027ad9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"ViT-B/16","Optimizer":"Lion","RandAug + Mixup":true,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 68.14. Result ID: 6990b0ff3a6c314849027ad8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"ViT-B/16","Optimizer":"AdamW","RandAug + Mixup":true,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 67.68. Result ID: 6990b0ff3a6c314849027ad5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"ViT-S/16","Optimizer":"Lion","RandAug + Mixup":true,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 66.73. Result ID: 6990b0ff3a6c314849027ad4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"ViT-S/16","Optimizer":"AdamW","RandAug + Mixup":true,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 64.81. Result ID: 6990b0ff3a6c314849027ad7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"ViT-B/16","Optimizer":"Lion","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 64.24. Result ID: 6990b0ff3a6c314849027acc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"ResNet-50","Optimizer":"AdamW","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 64.14. Result ID: 6990b0ff3a6c314849027ad3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"ViT-S/16","Optimizer":"Lion","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 64.02. Result ID: 6990b0ff3a6c314849027acd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"ResNet-50","Optimizer":"Lion","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### SGD Accuracy: 63.93. Result ID: 6990b0ff3a6c314849027acb. [Reporting paper]() Recorded evaluation configuration: {"Model":"ResNet-50","Optimizer":"SGD","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 63.09. Result ID: 6990b0ff3a6c314849027ad2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"ViT-S/16","Optimizer":"AdamW","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 61.87. Result ID: 6990b0ff3a6c314849027ad6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"ViT-B/16","Optimizer":"AdamW","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 55.94. Result ID: 6990b0ff3a6c314849027ad1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Mixer-B/16","Optimizer":"Lion","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### Lion Accuracy: 55.75. Result ID: 6990b0ff3a6c314849027acf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Mixer-S/16","Optimizer":"Lion","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 55.01. Result ID: 6990b0ff3a6c314849027ace. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Mixer-S/16","Optimizer":"AdamW","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. ### AdamW Accuracy: 53.37. Result ID: 6990b0ff3a6c314849027ad0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Mixer-B/16","Optimizer":"AdamW","RandAug + Mixup":false,"Training Protocol":"Train from scratch on ImageNet"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.