</> Code with Papers by Syntology

Datasets › CIFAR-100

CIFAR-100

Introduced in Learning multiple layers of features from tiny images8 Apr 2009 archive 2025-07-28

The CIFAR-100 dataset (Canadian Institute for Advanced Research, 100 classes) is a subset of the Tiny Images dataset and consists of 60000 32x32 color images. The 100 classes in the CIFAR-100 are grouped into 20 superclasses. There are 600 images per class. Each image comes with a "fine" label (the class to which it belongs) and a "coarse" label (the superclass to which it belongs). There are 500 training images and 100 testing images per class.

The criteria for deciding whether an image belongs to a class were as follows:

  • The class name should be high on the list of likely answers to the question “What is in this picture?”
  • The image should be photo-realistic. Labelers were instructed to reject line drawings.
  • The image should contain only one prominent instance of the object to which the class refers.
  • The object may be partially occluded or seen from an unusual viewpoint as long as its identity is still clear to the labeler.

Source: https://www.cs.toronto.edu/~kriz/cifar.html Image Source: https://www.cs.toronto.edu/~kriz/cifar.html

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 51 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Image Classification CIFAR-100 EffNet-L2 (SAM) Percentage correct 96.08 Sharpness-Aware Minimization for Efficiently Improving... davda54/sam +17 211 Compare
Long-tail Learning CIFAR-100-LT (ρ=100) LPT Error Rate 10.9 LPT: Long-tailed Prompt Tuning for Image Classification dongsky/lpt 66 Compare
Long-tail Learning CIFAR-100-LT (ρ=10) LIFT (ViT-B/16, ImageNet-21K pre-training) Error Rate 8.7 Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts shijxcs/lift 31 Compare
Image Clustering CIFAR-100 TURTLE (CLIP + DINOv2) Accuracy 0.898 Let Go of Your Labels with Unsupervised Transfer mlbio-epfl/turtle 30 Compare
Semi-Supervised Image Classification cifar-100, 10000 Labels Semi-SST (ViT-Small) Percentage error 13.50±0.14 SST: Self-training with Self-adaptive Thresholding for... — 29 Compare
Knowledge Distillation CIFAR-100 SRD (T:resnet-32x4, S:shufflenet-v2) Top-1 Accuracy (%) 79.86 Understanding the Role of the Projector in Knowledge Distillation yoshitomo-matsubara/torchdistill +3 27 Compare
Semi-Supervised Image Classification CIFAR-100, 400 Labels SemiReward Percentage error 15.62 SemiReward: A General Reward Model for Semi-supervised Learning Westlake-AI/SemiReward 21 Compare
Semi-Supervised Image Classification CIFAR-100, 2500 Labels Semi-SST (ViT-Small) Percentage error 16.62±0.28 SST: Self-training with Self-adaptive Thresholding for... — 16 Compare
Anomaly Detection One-class CIFAR-100 GeneralAD AUROC 98.4 GeneralAD: Anomaly Detection Across Domains by Attending... LucStrater/GeneralAD 15 Compare
Incremental Learning CIFAR-100 - 50 classes + 5 steps of 10 classes TCIL Average Incremental Accuracy 74.88 Resolving Task Confusion in Dynamic Expansion... yellowpancake/tcil 15 Compare
Anomaly Detection Unlabeled CIFAR-10 vs CIFAR-100 PsudoLabels ViT AUROC 96.7 Out-of-Distribution Detection Without Class Labels — 13 Compare
Incremental Learning CIFAR-100 - 50 classes + 10 steps of 5 classes TCIL Average Incremental Accuracy 73.72 Resolving Task Confusion in Dynamic Expansion... yellowpancake/tcil 13 Compare
Neural Architecture Search CIFAR-100 DNA-c Percentage Error 11.7 Blockwisely Supervised Neural Architecture Search with... changlin31/DNA 13 Compare
Few-Shot Class-Incremental Learning CIFAR-100 PriViLege Last Accuracy 86.06 Pre-trained Vision and Language Transformers Are... khu-agi/privilege 11 Compare
Continual Learning Cifar100 (20 tasks) Model Zoo-Continual Average Accuracy 94.99 Model Zoo: A Growing "Brain" That Learns Continually grasp-lyrl/modelzoo_continual +1 9 Compare
Image Generation CIFAR-100 LeCAM (StyleGAN2 + ADA) FID 2.99 Regularizing Generative Adversarial Networks under Limited Data google/lecam-gan 9 Compare
Class Incremental Learning cifar100 S&B 10-stage average accuracy 68.18 Split-and-Bridge: Adaptable Class Incremental Learning... bigdata-inha/Split-and-Bridge 7 Compare
Conditional Image Generation CIFAR-100 DLSM FID 3.86 Denoising Likelihood Score Matching for Conditional... chen-hao-chao/dlsm +1 7 Compare
Personalized Federated Learning CIFAR-100 pFedGP-IP-data ACC@1-500 55.7 Personalized Federated Learning with Gaussian Processes IdanAchituve/pFedGP 7 Compare
Incremental Learning CIFAR-100 - 50 classes + 25 steps of 2 classes D3Former Average Incremental Accuracy 68.68 D3Former: Debiased Dual Distilled Transformer for... abdohelmy/D-3Former 5 Compare
Network Pruning CIFAR-100 Dense Accuracy 79 AC/DC: Alternating Compressed/DeCompressed Training of... IST-DASLab/ACDC +1 5 Compare
Out-of-Distribution Detection CIFAR-100 Wide ResNet 40x2 FPR95 23.4 An Effective Baseline for Robustness to Distributional Shift Sushil-Thapa/Abstention-OoD 4 Compare
Provable Adversarial Defense CIFAR-100 SLL X-Large Accuracy 42.7 A Unified Algebraic Perspective on Lipschitz Neural Networks araujoalexandre/lipschitz-sll-networks 4 Compare
Adversarial Defense CIFAR-100 wideresnet-34-20 autoattack 62.55/30.20 Learnable Boundary Guided Adversarial Training fra31/auto-attack +2 3 Compare
Data Free Quantization CIFAR-100 ResNet-20 CIFAR-100 CIFAR-100 W4A4 Top-1 Accuracy 65.10 Qimera: Data-free Quantization with Synthetic Boundary... iamkanghyunchoi/qimera +1 3 Compare
Open-World Semi-Supervised Learning CIFAR-100 TRSSL (ResNet-18) All accuracy (10% Labeled) 60.3 Towards Realistic Semi-Supervised Learning nayeemrizve/trssl 3 Compare
Small Data Image Classification CIFAR-100, 1000 Labels ChimeraMix+AutoAugment Accuracy 35.02 ChimeraMix: Image Classification on Small Datasets via... creinders/chimeramix 3 Compare
Adversarial Attack CIFAR-100 3-ensemble of multi-resolution self-ensembles Attack: AutoAttack 51.28 Ensemble everything everywhere: Multi-scale aggregation... stanislavfort/ensemble-everything-everywhere +1 2 Compare
Adversarial Robustness CIFAR-100 Mixed Classifier Clean Accuracy 85.21 Improving the Accuracy-Robustness Trade-Off of... codelion/adaptive-classifier +1 2 Compare
Bayesian Inference cifar100 F-SWA Accuracy 83.61 — — 2 Compare
Class Incremental Learning CIFAR-100 - 50 classes + 10 steps of 5 classes PPCA-SWSL Final Accuracy 77.07 Scalable Learning with Incremental Probabilistic PCA barbua/PPCA 2 Compare
Class Incremental Learning CIFAR-100 - 50 classes + 5 steps of 10 classes PPCA-SWSL Final Accuracy 77.07 Scalable Learning with Incremental Probabilistic PCA barbua/PPCA 2 Compare
Few-Shot Image Classification CIFAR100 5-way (1-shot) UL-Hopfield (ULH) Accuracy 89.6 Unsupervised Learning using Pretrained CNN and... — 2 Compare
Incremental Learning CIFAR-100 - 50 classes + 50 steps of 1 class PODNet Average Incremental Accuracy 57.98 PODNet: Pooled Outputs Distillation for Small-Tasks... g-u-n/pycil +1 2 Compare
Learning with coarse labels cifar100 MaskCon Recall@1 65.52 MaskCon: Masked Contrastive Learning for Coarse-Labelled Dataset MrChenFeng/MaskCon_CVPR2023 2 Compare
Self-Supervised Learning cifar100 ResNet50 average top-1 classification accuracy 72.51 Guarding Barlow Twins Against Overfitting with Mixed Samples wgcban/mix-bt 2 Compare
Semi-Supervised Image Classification CIFAR-100, 5000Labels LiDAM Percentage correct 75.14 LiDAM: Semi-Supervised Learning with Localized Domain... — 2 Compare
Stochastic Optimization CIFAR-100 Resnet18 Accuracy (max) 58.48 Mixing ADAM and SGD: a Combined Optimization Method gitlab.com/nicolalandro/multi_optimizer 2 Compare
Zero-Shot Learning CIFAR-100 ZLaP* Accuracy 74.2 Label Propagation for Zero-shot Classification with... vladan-stojnic/zlap 2 Compare
class-incremental learning cifar100 EWC 10-stage average accuracy 50.53 Overcoming catastrophic forgetting in neural networks ContinualAI/avalanche +28 1 Compare
Classification CIFAR-100 ResNet8×4 Accuracy 77.50 LumiNet: The Bright Side of Perceptual Knowledge Distillation ismail31416/luminet 1 Compare
Classifier calibration CIFAR-100 R-Mix (PreActResNet-18) Expected Calibration Error 3.73 Expeditious Saliency-guided Mix-up through Random... minhlong94/random-mixup 1 Compare
Image Classification cifar100 shreynet 1:1 Accuracy 45.98 Deep Residual Learning for Image Recognition tensorflow/models +483 1 Compare
Learning with noisy labels CIFAR-100 InstanceGM Test Accuracy 77.19 Instance-Dependent Noisy Label Learning via Graphical Modelling arpit2412/InstanceGM 1 Compare
Non-exemplar-based Class Incremental Learning cifar100 NAPA-VQ Average accuracy - 5 tasks 70.44 NAPA-VQ: Neighborhood Aware Prototype Augmentation with... tamasham/napa-vq 1 Compare
Novel Class Discovery cifar100 AutoNovel Clustering Accuracy 0.746 AutoNovel: Automatically Discovering and Learning Novel... k-han/AutoNovel 1 Compare
Out-of-Distribution Detection cifar100 Wide Resnet 40x2 AUROC 95.76 RODD: A Self-Supervised Approach for Robust... UmarKhalidcs/RODD 1 Compare
Self-Supervised Learning CIFAR-100 CorInfomax (ResNet18) Top-1 Accuracy 71.61 Self-Supervised Learning with an Information... serdarozsoy/corinfomax-ssl 1 Compare
Semi-Supervised Image Classification CIFAR-100, 1000 Labels EnAET Percentage correct 41.27 EnAET: A Self-Trained framework for Semi-Supervised and... maple-research-lab/EnAET +1 1 Compare
Transductive Zero-Shot Classification CIFAR-100 ZLaP Accuarcy 73.3 Label Propagation for Zero-shot Classification with... vladan-stojnic/zlap 1 Compare
Classification cifar100 no rows — — 0 Compare

Papers archive 2025-07-28

30 shown of 375 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 9,045. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Linear Attention with Global Context: A Multipole Attention Mechanism for Vision and Physics 1 1 3 Jul 2025 not harvested
ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using Sparse Labels 1 2 4 Jun 2025 not harvested
SST: Self-training with Self-adaptive Thresholding for Semi-supervised Learning 0 6 31 May 2025 not harvested
Exploring a Principled Framework for Deep Subspace Clustering 1 1 21 Mar 2025 ran 4 of 4 samples (0 unverified; 4 pointer-only for licence)
Label Ranker: Self-Aware Preference for Classification Label Position in Visual Masked Self-Supervised Pre-Trained Model 1 1 3 Mar 2025 not harvested
Deep Clustering via Probabilistic Ratio-Cut Optimization 1 1 1 Feb 2025 not harvested
How transfer learning is used in generative models for image classification: improved accuracy 1 1 9 Dec 2024 not harvested
On the Performance Analysis of Momentum Method: A Frequency Domain Perspective 1 1 29 Nov 2024 ran 1 of 2 samples (1 unverified)
ANDHRA Bandersnatch: Training Neural Networks to Predict Parallel Realities 1 5 28 Nov 2024 not harvested
Deep Feature Response Discriminative Calibration 1 1 16 Nov 2024 not harvested
Enhancing GANs with MMD Neural Architecture Search, PMish Activation Function, and Adaptive Rank Decomposition 1 2 23 Oct 2024 not harvested
Performance of Gaussian Mixture Model Classifiers on Embedded Feature Spaces 1 1 17 Oct 2024 not harvested
Improving Image Clustering with Artifacts Attenuation via Inference-Time Attention Engineering 0 1 7 Oct 2024 not harvested
Stochastic Subsampling With Average Pooling 0 1 25 Sep 2024 not harvested
Ensemble everything everywhere: Multi-scale aggregation for adversarial robustness 2 2 8 Aug 2024 ran 2 of 3 samples (1 unverified; 3 pointer-only for licence)
GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Features 1 1 17 Jul 2024 ran 3 of 4 samples (1 unverified)
Deep Online Probability Aggregation Clustering 1 1 7 Jul 2024 ran 5 of 10 samples (5 unverified; 10 pointer-only for licence)
Enhanced Long-Tailed Recognition with Contrastive CutMix Augmentation 2 1 6 Jul 2024 not harvested
The Balanced-Pairwise-Affinities Feature Transform 1 1 25 Jun 2024 not harvested
Let Go of Your Labels with Unsupervised Transfer 1 1 11 Jun 2024 ran 3 of 4 samples (1 unverified; 4 pointer-only for licence)
Few-shot Tuning of Foundation Models for Class-incremental Learning 1 1 26 May 2024 ran 6 of 10 samples (4 unverified)
Harnessing Hierarchical Label Distribution Variations in Test Agnostic Long-tail Recognition 1 1 13 May 2024 ran 1 of 1 samples (0 unverified; 1 pointer-only for licence)
Label Propagation for Zero-shot Classification with Vision-Language Models 1 3 5 Apr 2024 ran 1 of 2 samples (1 unverified)
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets 2 1 3 Apr 2024 not harvested
Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners 1 1 2 Apr 2024 ran 12 of 14 samples (2 unverified)
A Bag of Tricks for Few-Shot Class-Incremental Learning 0 1 21 Mar 2024 not harvested
Logit Standardization in Knowledge Distillation 1 2 3 Mar 2024 ran 2 of 2 samples (0 unverified; 2 pointer-only for licence)
SURE: SUrvey REcipes for building reliable and robust deep networks 1 2 1 Mar 2024 ran 5 of 6 samples (1 unverified; 6 pointer-only for licence)
Pre-training of Lightweight Vision Transformers on Small Datasets with Minimally Scaled Images 0 1 6 Feb 2024 not harvested
Guarding Barlow Twins Against Overfitting with Mixed Samples 1 2 4 Dec 2023 not harvested

The full list of 375 is in the JSON twin.

Dataset loaders archive 2025-07-28

huggingface.co/datasets/uoft-cs/cifar100huggingface/datasetstfpytorchjax
huggingface.co/datasets/cifar100huggingface/datasetstfpytorchjax
huggingface.co/datasets/tomas-gajarsky/cifar100-lthuggingface/datasetstfpytorchjax
huggingface.co/datasets/renumics/cifar100-enrichedhuggingface/datasetstfpytorchjax
huggingface.co/datasets/renumics/cifar100-outlierhuggingface/datasetstfpytorchjax
huggingface.co/datasets/wltjr1007/cifar100_cliphuggingface/datasetstfpytorchjax
pytorch.org/vision/stable/generated/torchvision.datasets.CIFAR100.htmlpytorch/visionpytorch
docs.voxel51.com/user_guide/dataset_zoo/datasets.html#cifar-100voxel51/fiftyonetfpytorch
datasets.activeloop.ai/docs/ml/datasets/cifar-100-dataset/activeloopai/deeplaketfpytorch
www.tensorflow.org/datasets/catalog/cifar100tensorflow/datasetstfjax
open-mmlab/mmclassificationopen-mmlab/mmclassificationpytorch

11 loaders as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

Image ClassificationClassificationImage GenerationZero-Shot LearningAnomaly DetectionFew-Shot Image ClassificationSemi-Supervised Image ClassificationContinual LearningImage ClusteringOut-of-Distribution DetectionNeural Architecture SearchClass Incremental LearningLong-tail LearningTransductive Zero-Shot ClassificationIncremental LearningLearning with noisy labelsBinarizationImage Classification with Label NoiseConditional Image GenerationStochastic OptimizationAdversarial DefenseSmall Data Image ClassificationKnowledge DistillationSelf-Supervised LearningPersonalized Federated LearningAdversarial RobustnessNetwork PruningLearning with coarse labelsDataset Distillation - 1IPCAdversarial AttackClassification with Binary Weight NetworkClassification with Binary Neural NetworkSparse Learning and binarizationNovel Class DiscoveryOpen-World Semi-Supervised LearningFew-Shot Class-Incremental LearningNon-exemplar-based Class Incremental LearningProvable Adversarial DefenseData Free QuantizationBayesian InferenceVariational InferenceClassifier calibrationclass-incremental learning

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

Images

Languages archive 2025-07-28

Chinese

Variants archive 2025-07-28

  • cifar100
  • Unlabeled CIFAR-10 vs CIFAR-100
  • CIFAR-100 ResNet-18 - 200 Epochs
  • CIFAR-10, 2000 Labeled Samples
  • cifar-100, 10000 Labels
  • One-class CIFAR-100
  • Cifar100 (20 tasks)
  • CIFAR100 5-way (1-shot)
  • CIFAR-100-LT (ρ=100)
  • CIFAR-100-LT (ρ=10)
  • CIFAR-100, 5000Labels
  • CIFAR-100, 400 Labels
  • CIFAR-100, 2500 Labels
  • CIFAR-100, 1000 Labels
  • CIFAR-100 - 50 classes + 50 steps of 1 class
  • CIFAR-100 - 50 classes + 5 steps of 10 classes
  • CIFAR-100 - 50 classes + 25 steps of 2 classes
  • CIFAR-100 - 50 classes + 10 steps of 5 classes
  • CIFAR-100

19 variant names, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections