CIFAR-100

Unknown

60K 32x32 color images in 100 fine-grained classes grouped into 20 superclasses. More challenging than CIFAR-10.

Benchmark Stats

Models4
Papers4
Metrics1

SOTA History

Coming Soon
Visual timeline of state-of-the-art progression over time will appear here.

accuracy

accuracy

Higher is better

RankModelCodeScorePaper / Source
1vit-h-14

Fine-tuned from ImageNet pretraining.

94.55google-research
2efficientnet-b7

Transfer learning from ImageNet.

91.7google-research
3vit-b-16

Fine-tuned from ImageNet-21K.

91.48huggingface
4resnet-50

With Cutout augmentation.

78.04cutout-paper