Leaderboard Explore Compare Profile Community Blog FAQ Competition 2024 Report Issue

Sample stimuli

How to use

from brainscore_vision import load_benchmark
benchmark = load_benchmark("Kato2026Dense-accuracy_distance")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	eva02_large_in22k	.854
2	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.848
3	convnext_xxlarge_clip	.799
4	alexnet_training_seed_03	.764
5	vit_huge_clip336_laion	.644
6	dinov2_vitl14_lc	.636
7	vit_huge_patch14_clip_336:laion2b_ft_in12k_in1k	.613
8	alexnet_training_seed_09	.611
9	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.608
10	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.603
11	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.597
12	dinov2_vitg14_lc	.593
13	dinov2_vitg14_reg_lc	.587
14	vit_large_patch14_clip_336:laion2b_ft_in1k	.576
15	convnext_large_clip_laion2b_finelayers	.573
16	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.555
17	dinov2_vitl14_lc_res336	.554
18	dinov2_vitl14_reg_lc	.552
19	mehrer2020_alexnet_training_seed_01_fov4	.543
20	vit_large_clip336_openai	.539
21	vit_large_patch14_clip_224:laion2b_ft_in1k	.520
22	alexnet_training_seed_07	.476
23	vit_large_patch14_clip_224:openai_ft_in1k	.471
24	AdvProp_efficientnet-b7	.389
25	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.368
26	resnext101_32x48d_wsl	.358
27	resnext101_32x32d_wsl	.342
28	AdvProp_efficientnet-b8	.293
29	AdvProp_efficientnet-b4	.291
30	convnext_xlarge:fb_in22k_ft_in1k	.286
31	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.210
32	AdvProp_efficientnet-b6	.202
33	convnext_large:fb_in22k_ft_in1k	.192
34	AdvProp_efficientnet-b2	.192
35	AT_efficientnet-b2	.192
36	convnext_base:clip_laiona_augreg_ft_in1k_384	.188
37	densenet-201	.186
38	convnext_base_fb_in22k_ft_in1k	.173
39	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.165
40	eicircuit_resnet18_bn_20260919	.163
41	convnext_large_in22k_supervised	.150
42	resnext101_32x16d_wsl	.128
43	ReAlnet07	.121
44	ReAlnet10	.121
45	ReAlnet04	.119
46	resnet50_byol_linhead	.117
47	mehrer2020_alexnet_training_seed_01_fov16	.109
48	resnet50-SIN	.109
49	convnext_large_imagenet_full_seed-0	.109
50	deit_large_imagenet_full_seed-0	.103
51	shufflenet_v2_x1_0	.102
52	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.100
53	resnet50_vicreg_linhead	.100
54	resnet34_imagenet_full	.095
55	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.093
56	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.084
57	antialias-resnet152	.081
58	resnet50-moclr8deg	.081
59	resnet50_robust_l2_eps3	.074
60	imagenet_l2_3_0	.074
61	resnet-50-robust	.074
62	resnet_101_v2	.072
63	ReAlnet09	.072
64	ReAlnet08	.070
65	pnasnet_large_halfPrecision	.064
66	resnet_50_v1_spiking_l4a	.063
67	resnet50_barlowtwins_linhead	.061
68	mehrer2020_alexnet_training_seed_05	.061
69	deit3_base_patch16_224_fb_in22k_ft_in1k	.060
70	resnext101_32x8d_wsl	.055
71	cvt_cvt-21-384-in1k_4_LucyV4	.054
72	mehrer2020_alexnet_training_seed_03	.052
73	convnext_femto_ols:d1_in1k	.052
74	convnext_base_fb_in1k	.050
75	alexnet	.049
76	resnet50-SIN_IN	.049
77	regnety_032	.047
78	grcnn	.046
79	grcnn_109	.045
80	resnet50_ewconplus_wzc_seed2_2__323	.044
81	resnet_101_v1	.042
82	resnet50_hvs_wzc	.041
83	AlexNet_SIN_fov12	.041
84	mehrer2020_alexnet_training_seed_04	.037
85	mehrer2020_alexnet_training_seed_02	.035
86	vit_base_patch16_clip_224:openai_ft_in1k	.033
87	resnet50_simclr_linhead_01norm	.033
88	mobilenet_v2_1_4_224	.031
89	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.031
90	focalnet_tiny_lrf_in1k	.029
91	vit_relpos_base_patch16_clsgap_224:sw_in1k	.028
92	convnext_small_imagenet_100_seed-0	.024
93	mehrer2020_alexnet_training_seed_01	.020
94	convnext_base_imagenet_full_seed-0	.020
95	densenet-121	.019
96	resnet_50_v1	.018
97	resnet50_tutorial	.018
98	resnet50-sup	.018
99	convnext_tiny_sup	.018
100	resnet_152_v2	.017
101	cvt_cvt-21-224-in1k_4_LucyV4	.017
102	deit3_base_patch16_224_fb_in1k	.015
103	convnext_base_fb_in1k_gram_aligned_dinov3	.015
104	nasnet_mobile	.014
105	resnet50_st_wzc_seed3_2_323	.012
106	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.012
107	resnet50_robust_l2_eps1	.012
108	resnet50_simsiam_linhead	.000
109	ijepa_vith14_linhead	.000
110	mehrer2020_alexnet_training_seed_01_fov12	.000
111	voneresnet-50-robust	.000
112	vonegrcnn_62e_nobn	.000
113	vonegrcnn_47e	.000
114	vgg_16	.000
115	seresnext50_32x4d	.000
116	resnet50_st_wzc_seed7_2_323	.000
117	resnet50_st_wzc_seed6_2_323	.000
118	resnet50_st_wzc_2	.000
119	resnet50_ewconplus_wzc_seed5_2__323	.000
120	resnet50_ewconplus_wzc_seed6_2_323	.000
121	resnet18_imagenet_full	.000
122	resnet101_imagenet_full	.000
123	efficientnet_b0_imagenet_full	.000
124	inception_v3	.000
125	efficientnet_b1_imagenet_full	.000
126	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.000
127	cv_18_dagger_408_pretrained	.000
128	blt_vs	.000
129	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.000
130	voneresnet-50-non_stochastic	.000
131	voneresnet-50	.000
132	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.000
133	resnet_50_v1_spiking	.000
134	resnet50_ewconplus_wzc	.000
135	resnet_SIN_IN_FT_IN	.000
136	resnet_50_v2	.000
137	resnet50_imagenet_full	.000
138	resnet50-vitoimagevidnet8	.000
139	resnet50-SIN_IN_IN	.000
140	resnet50-VITO-8deg-cc	.000
141	resnet152_imagenet_full	.000
142	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.000
143	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.000
144	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.000
145	effnetb1_272x240	.000
146	efficientnet_b2_imagenet_full	.000
147	efficientnet_b0	.000
148	deit_base_imagenet_full_seed-0	.000
149	deit_small_imagenet_full_seed-0	.000
150	cvt_cvt-13-384-in1k_4_LucyV4	.000
151	cvt_cvt-13-224-in1k_4_LucyV4	.000
152	custom_model_cv_18_dagger_408	.000
153	convnext_tiny_imagenet_full_seed-0	.000
154	convnext_small_imagenet_full_seed-0	.000
155	convnext_tiny:in12k_ft_in1k	.000
156	antialiased-r50	.000
157	ViT_L_32_imagenet1k	.000
158	Res2Net50_26w_4s	.000
159	AlexNet_SIN_fov16	.000
160	AlexNet_SIN	.000
161	dinov2_vitl14_reg_lc_featread
162	mobilenet_v2_1_0_128
163	gabor_filter_entropy_4x4_v2_forcedCosine
164	gabor_filter_entropy_4x4_v2
165	deepeisnn_ei_imagenet_resnet18
166	alexnet_training_seed_10
167	alexnet_training_seed_06
168	alexnet_training_seed_02
169	alexnet_training_seed_01
170	hmax
171	ReAlnet01_cornet_fov4
172	ReAlnet01_cornet
173	compact_resnet50_robust_V4
174	VOneCORnet-S
175	ReAlnet03_cornet
176	ReAlnet04_cornet

Benchmark bibtex

None

Ceiling

0.82.

Note that scores are relative to this ceiling.

Data: Kato2026Dense

Metric: accuracy_distance