Image Classification on ImageNet

Metric: Top 1 Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Top 1 Accuracy▼	Extra Data	Paper	Date↕	Code
1	Unicom (ViT-L/14@336px) (Finetuned)	88.3	No	Unicom: Universal and Compact Representation Lea...	2023-04-12	Code
2	Bamboo (Bamboo-H)	87.1	No	A Study on Transformer Configuration and Trainin...	2022-05-21	-
3	DINOv2+reg (ViT-g/14)	87.1	Yes	Vision Transformers Need Registers	2023-09-28	Code
4	Bamboo (Bamboo-L)	86.3	No	A Study on Transformer Configuration and Trainin...	2022-05-21	-
5	TinySaver(ConvNeXtV2_h, 0.01 Acc drop)	86.24	No	Tiny Models are the Computational Saver for Larg...	2024-03-26	Code
6	Refiner-ViT-L	86.03	No	Refiner: Refining Self-attention for Vision Tran...	2021-06-07	Code
7	TinySaver(ConvNeXtV2_h, 0.5 Acc drop)	85.75	No	Tiny Models are the Computational Saver for Larg...	2024-03-26	Code
8	TinySaver(Swin_large, 0.5 Acc drop)	85.74	No	Tiny Models are the Computational Saver for Larg...	2024-03-26	Code
9	TinySaver(Swin_large, 1.0 Acc drop)	85.24	No	Tiny Models are the Computational Saver for Larg...	2024-03-26	Code
10	Bamboo (Bamboo-B)	84.2	No	A Study on Transformer Configuration and Trainin...	2022-05-21	-
11	AIM-7B	84	No	Scalable Pre-training of Large Autoregressive Im...	2024-01-16	Code
12	DynamicViT-LV-M/0.8	83.9	No	DynamicViT: Efficient Vision Transformers with D...	2021-06-03	Code
13	TinySaver(EfficientFormerV2_l, 0.01 Acc drop)	83.52	No	Tiny Models are the Computational Saver for Larg...	2024-03-26	Code
14	KAT-B*	82.8	No	Kolmogorov-Arnold Transformer	2024-09-16	Code
15	ReViT-B	82.4	No	ReViT: Enhancing Vision Transformers Feature Div...	2024-02-17	Code
16	ConvNeXt-T-Hermite	82.34	No	Polynomial, trigonometric, and tropical activati...	2025-02-03	Code
17	ConvMixer-1536/20	82.2	No	Patches Are All You Need?	2022-01-24	Code
18	DIFFQ (λ=1e−2)	82	No	Differentiable Model Compression via Pseudo Quan...	2021-04-20	Code
19	DeiT-B	81.8	No	Kolmogorov-Arnold Transformer	2024-09-16	Code
20	EsViT (Swin-B)	81.3	No	Efficient Self-supervised Vision Transformers fo...	2021-06-17	Code
21	SimpleNetV1-9m-correct-labels	81.24	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
22	ResNeXt-101 (Debiased+CutMix)	81.2	No	Shape-Texture Debiased Neural Network Training	2020-10-12	Code
23	EsViT(Swin-S)	80.8	No	Efficient Self-supervised Vision Transformers fo...	2021-06-17	Code
24	SimpleNetV1-5m-correct-labels	79.12	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
25	ViT-B/16	79.1	No	Kolmogorov-Arnold Transformer	2024-09-16	Code
26	Inception V3	78.8	No	-	-	-
27	CSAT	78.6	No	-	-	-
28	ConvMLP-S	76.8	No	ConvMLP: Hierarchical Convolutional MLPs for Vis...	2021-09-09	Code
29	VGG	76.3	Yes	-	-	-
30	ELP (naive ResNet50)	76.13	No	-	-	Code
31	SimpleNetV1-small-075-correct-labels	75.66	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
32	FF	74.9	No	Do You Even Need Attention? A Stack of Feed-Forw...	2021-05-06	Code
33	SimpleNetV1-9m	74.17	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
34	VICReg (ResNet50)	73.2	No	VICReg: Variance-Invariance-Covariance Regulariz...	2021-05-11	Code
35	I-VNE+ (ResNet-50)	72.1	No	VNE: An Effective Method for Improving Deep Repr...	2023-04-04	Code
36	SimpleNetV1-5m	71.94	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
37	Dspike (VGG-16)	71.24	Yes	-	-	-
38	PSN (SEW ResNet-34)	70.54	No	-	-	-
39	GAC-SNN MS-ResNet-34	70.42	No	Gated Attention Coding for Training High-perform...	2023-08-12	Code
40	SimpleNetV1-small-05-correct-labels	69.11	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
41	SimpleNetV1-small-075	68.15	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
42	PSN (SEW ResNet-18)	67.63	No	-	-	-
43	OverFeat	66.04	Yes	-	-	-
44	DGPPF-ResNet18	65.22	No	-	-	Code
45	Alexnet	63.3	Yes	-	-	-
46	SimpleNetV1-small-05	61.52	No	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
47	DeepCluster (AlexNet)	41	No	Deep Clustering for Unsupervised Learning of Vis...	2018-07-15	Code
48	Colorisation (improved) (ResNet-101)	39.6	No	Multi-task Self-Supervised Visual Learning	2017-08-25	-
49	NFResnet-50	39.2	No	TAN Without a Burn: Scaling Laws of DP-SGD	2022-10-07	Code
50	Rotation (AlexNet)	38.7	No	Unsupervised Representation Learning by Predicti...	2018-03-21	Code
51	Counting (AlexNet)	34.3	No	Representation Learning by Learning to Count	2017-08-22	Code
52	NFResnet-50	32.4	No	Unlocking High-Accuracy Differentially Private I...	2022-04-28	Code
53	Resnet-18	6.9	No	Toward Training at ImageNet Scale with Different...	2022-01-28	Code
54	Resnet-50	5	No	Toward Training at ImageNet Scale with Different...	2022-01-28	Code

#1Unicom (ViT-L/14@336px) (Finetuned)SOTA
88.3
Top 1 Accuracy· 2023-04-12
Unicom: Universal and Compact Representation Learning for Image Retrieval Code
#2Bamboo (Bamboo-H)SOTA
87.1
Top 1 Accuracy· 2022-05-21
A Study on Transformer Configuration and Training Objective
#3DINOv2+reg (ViT-g/14)
87.1
Top 1 Accuracy· Extra Data· 2023-09-28
Vision Transformers Need Registers Code
#4Bamboo (Bamboo-L)
86.3
Top 1 Accuracy· 2022-05-21
A Study on Transformer Configuration and Training Objective
#5TinySaver(ConvNeXtV2_h, 0.01 Acc drop)
86.24
Top 1 Accuracy· 2024-03-26
Tiny Models are the Computational Saver for Large Models Code
#6Refiner-ViT-LSOTA
86.03
Top 1 Accuracy· 2021-06-07
Refiner: Refining Self-attention for Vision Transformers Code
#7TinySaver(ConvNeXtV2_h, 0.5 Acc drop)
85.75
Top 1 Accuracy· 2024-03-26
Tiny Models are the Computational Saver for Large Models Code
#8TinySaver(Swin_large, 0.5 Acc drop)
85.74
Top 1 Accuracy· 2024-03-26
Tiny Models are the Computational Saver for Large Models Code
#9TinySaver(Swin_large, 1.0 Acc drop)
85.24
Top 1 Accuracy· 2024-03-26
Tiny Models are the Computational Saver for Large Models Code
#10Bamboo (Bamboo-B)
84.2
Top 1 Accuracy· 2022-05-21
A Study on Transformer Configuration and Training Objective
#11AIM-7B
84
Top 1 Accuracy· 2024-01-16
Scalable Pre-training of Large Autoregressive Image Models Code
#12DynamicViT-LV-M/0.8SOTA
83.9
Top 1 Accuracy· 2021-06-03
DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification Code
#13TinySaver(EfficientFormerV2_l, 0.01 Acc drop)
83.52
Top 1 Accuracy· 2024-03-26
Tiny Models are the Computational Saver for Large Models Code
#14KAT-B*
82.8
Top 1 Accuracy· 2024-09-16
Kolmogorov-Arnold Transformer Code
#15ReViT-B
82.4
Top 1 Accuracy· 2024-02-17
ReViT: Enhancing Vision Transformers Feature Diversity with Attention Residual Connections Code
#16ConvNeXt-T-Hermite
82.34
Top 1 Accuracy· 2025-02-03
Polynomial, trigonometric, and tropical activations Code
#17ConvMixer-1536/20
82.2
Top 1 Accuracy· 2022-01-24
Patches Are All You Need?Code
#18DIFFQ (λ=1e−2)SOTA
82
Top 1 Accuracy· 2021-04-20
Differentiable Model Compression via Pseudo Quantization Noise Code
#19DeiT-B
81.8
Top 1 Accuracy· 2024-09-16
Kolmogorov-Arnold Transformer Code
#20EsViT (Swin-B)
81.3
Top 1 Accuracy· 2021-06-17
Efficient Self-supervised Vision Transformers for Representation Learning Code
#21SimpleNetV1-9m-correct-labelsSOTA
81.24
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#22ResNeXt-101 (Debiased+CutMix)
81.2
Top 1 Accuracy· 2020-10-12
Shape-Texture Debiased Neural Network Training Code
#23EsViT(Swin-S)
80.8
Top 1 Accuracy· 2021-06-17
Efficient Self-supervised Vision Transformers for Representation Learning Code
#24SimpleNetV1-5m-correct-labels
79.12
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#25ViT-B/16
79.1
Top 1 Accuracy· 2024-09-16
Kolmogorov-Arnold Transformer Code
#26Inception V3
78.8
Top 1 Accuracy
No paper
#27CSAT
78.6
Top 1 Accuracy
No paper
#28ConvMLP-S
76.8
Top 1 Accuracy· 2021-09-09
ConvMLP: Hierarchical Convolutional MLPs for Vision Code
#29VGG
76.3
Top 1 Accuracy· Extra Data
No paper
#30ELP (naive ResNet50)
76.13
Top 1 Accuracy
No paperCode
#31SimpleNetV1-small-075-correct-labels
75.66
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#32FF
74.9
Top 1 Accuracy· 2021-05-06
Do You Even Need Attention? A Stack of Feed-Forward Layers Does Surprisingly Well on ImageNet Code
#33SimpleNetV1-9m
74.17
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#34VICReg (ResNet50)
73.2
Top 1 Accuracy· 2021-05-11
VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning Code
#35I-VNE+ (ResNet-50)
72.1
Top 1 Accuracy· 2023-04-04
VNE: An Effective Method for Improving Deep Representation by Manipulating Eigenvalue Distribution Code
#36SimpleNetV1-5m
71.94
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#37Dspike (VGG-16)
71.24
Top 1 Accuracy· Extra Data
No paper
#38PSN (SEW ResNet-34)
70.54
Top 1 Accuracy
No paper
#39GAC-SNN MS-ResNet-34
70.42
Top 1 Accuracy· 2023-08-12
Gated Attention Coding for Training High-performance and Efficient Spiking Neural Networks Code
#40SimpleNetV1-small-05-correct-labels
69.11
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#41SimpleNetV1-small-075
68.15
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#42PSN (SEW ResNet-18)
67.63
Top 1 Accuracy
No paper
#43OverFeat
66.04
Top 1 Accuracy· Extra Data
No paper
#44DGPPF-ResNet18
65.22
Top 1 Accuracy
No paperCode
#45Alexnet
63.3
Top 1 Accuracy· Extra Data
No paper
#46SimpleNetV1-small-05
61.52
Top 1 Accuracy· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#47DeepCluster (AlexNet)
41
Top 1 Accuracy· 2018-07-15
Deep Clustering for Unsupervised Learning of Visual Features Code
#48Colorisation (improved) (ResNet-101)
39.6
Top 1 Accuracy· 2017-08-25
Multi-task Self-Supervised Visual Learning
#49NFResnet-50
39.2
Top 1 Accuracy· 2022-10-07
TAN Without a Burn: Scaling Laws of DP-SGD Code
#50Rotation (AlexNet)
38.7
Top 1 Accuracy· 2018-03-21
Unsupervised Representation Learning by Predicting Image Rotations Code
#51Counting (AlexNet)
34.3
Top 1 Accuracy· 2017-08-22
Representation Learning by Learning to Count Code
#52NFResnet-50
32.4
Top 1 Accuracy· 2022-04-28
Unlocking High-Accuracy Differentially Private Image Classification through Scale Code
#53Resnet-18
6.9
Top 1 Accuracy· 2022-01-28
Toward Training at ImageNet Scale with Differential Privacy Code
#54Resnet-50
5
Top 1 Accuracy· 2022-01-28
Toward Training at ImageNet Scale with Differential Privacy Code