CIFAR-100
Unknown
60K 32x32 color images in 100 fine-grained classes grouped into 20 superclasses. More challenging than CIFAR-10.
Benchmark Stats
SOTA History
accuracy
accuracy
Higher is better
| Rank | Model | Source | Score | Year | Paper |
|---|---|---|---|---|---|
| 1 | EVA-02-L EVA-02-L/14+ fine-tuned on CIFAR-100. Pre-trained with EVA-CLIP on Objects365 + ImageNet-21K. State-of-the-art as of 2023. arXiv Mar 2023. | Editorial | 97.15 | 2026 | Source |
| 2 | CoAtNet-7 CoAtNet-7 (2.4B params) fine-tuned on CIFAR-100. Pre-trained on ImageNet-21K. arXiv Jun 2021, NeurIPS 2021. | Editorial | 96.38 | 2026 | Source |
| 3 | ConvNeXt V2-H ConvNeXt V2-H fine-tuned on CIFAR-100 after FCMAE pre-training on ImageNet-22K. arXiv Jan 2023, CVPR 2023. | Editorial | 96.17 | 2026 | Source |
| 4 | MAE ViT-H/14 ViT-H/14 fine-tuned on CIFAR-100 after MAE pre-training on ImageNet-1K. arXiv Nov 2021, CVPR 2022. | Editorial | 96.08 | 2026 | Source |
| 5 | SwinV2-G SwinV2-G (3B params) fine-tuned on CIFAR-100. Pre-trained on ImageNet-21K with resolution 192^2. arXiv Nov 2021, CVPR 2022. | Editorial | 96.01 | 2026 | Source |
| 6 | DeiT III-H/14 DeiT III ViT-H/14 fine-tuned on CIFAR-100. Improved training recipe for ViTs. arXiv Apr 2022, ECCV 2022. | Editorial | 95.94 | 2026 | Source |
| 7 | InternImage-XL InternImage-XL fine-tuned on CIFAR-100. Uses deformable convolutions as core operator. arXiv Nov 2022, CVPR 2023. | Editorial | 95.77 | 2026 | Source |
| 8 | FasterViT-6 FasterViT-6 fine-tuned on CIFAR-100. Hierarchical ViT with carrier tokens for high-resolution efficiency. arXiv Jun 2023. | Editorial | 95.72 | 2026 | Source |
| 9 | vit-h-14 Fine-tuned from ImageNet pretraining. | Editorial | 94.55 | 2025 | Source |
| 10 | ViT-L/16 (IN-21K) Vision Transformer ViT-L/16, pretrained on ImageNet-21K and finetuned on CIFAR-100. 93.25% reported in ViT paper (Table 5). Paper: Dosovitskiy et al. 2021, arxiv:2010.11929. | Community | 93.25 | 2026 | Source |
| 11 | efficientnet-b7 Transfer learning from ImageNet. | Editorial | 91.7 | 2025 | Source |
| 12 | vit-b-16 Fine-tuned from ImageNet-21K. | Editorial | 91.48 | 2025 | Source |
| 13 | resnet-50 With Cutout augmentation. | Editorial | 78.04 | 2025 | Source |