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("Kato2026Lin-error_consistency")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	convnext_xxlarge_clip	.914
2	vit_huge_patch14_clip_336:laion2b_ft_in12k_in1k	.910
3	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.900
4	vit_huge_clip336_laion	.883
5	convnext_large_clip_laion2b_finelayers	.858
6	dinov2_vitg14_lc	.829
7	eva02_large_in22k	.828
8	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.827
9	dinov2_vitl14_reg_lc	.814
10	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.812
11	convnext_base_fb_in22k_ft_in1k	.777
12	AdvProp_efficientnet-b4	.772
13	vit_large_patch14_clip_224:laion2b_ft_in1k	.771
14	dinov2_vitg14_reg_lc	.732
15	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.729
16	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.723
17	resnext101_32x48d_wsl	.723
18	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.720
19	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.717
20	vit_large_patch14_clip_336:laion2b_ft_in1k	.705
21	convnext_large:fb_in22k_ft_in1k	.701
22	resnext101_32x32d_wsl	.700
23	dinov2_vitl14_lc	.697
24	dinov2_vitl14_lc_res336	.664
25	convnext_large_in22k_supervised	.664
26	resnext101_32x16d_wsl	.640
27	vit_large_patch14_clip_224:openai_ft_in1k	.630
28	convnext_base:clip_laiona_augreg_ft_in1k_384	.622
29	deit3_base_patch16_224_fb_in22k_ft_in1k	.619
30	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.599
31	AdvProp_efficientnet-b6	.589
32	AdvProp_efficientnet-b7	.587
33	convnext_xlarge:fb_in22k_ft_in1k	.575
34	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.564
35	vit_base_patch16_clip_224:openai_ft_in1k	.560
36	convnext_base_fb_in1k	.554
37	regnety_032	.549
38	resnext101_32x8d_wsl	.537
39	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.530
40	vit_large_clip336_openai	.515
41	AT_efficientnet-b2	.515
42	AdvProp_efficientnet-b2	.515
43	deit3_base_patch16_224_fb_in1k	.507
44	AdvProp_efficientnet-b8	.504
45	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.499
46	focalnet_tiny_lrf_in1k	.493
47	convnext_tiny_imagenet_full_seed-0	.487
48	convnext_small_imagenet_full_seed-0	.483
49	convnext_base_imagenet_full_seed-0	.479
50	convnext_tiny:in12k_ft_in1k	.470
51	ViT_L_32_imagenet1k	.467
52	resnet50_st_wzc_2	.443
53	seresnext50_32x4d	.440
54	convnext_tiny_sup	.434
55	antialias-resnet152	.431
56	cv_18_dagger_408_pretrained	.428
57	convnext_large_imagenet_full_seed-0	.424
58	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.414
59	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.408
60	vit_relpos_base_patch16_clsgap_224:sw_in1k	.404
61	grcnn_109	.397
62	resnet_101_v2	.393
63	efficientnet_b0	.392
64	AlexNet_SIN_fov16	.391
65	resnet_101_v1	.388
66	deit_large_imagenet_full_seed-0	.388
67	convnext_base_fb_in1k_gram_aligned_dinov3	.380
68	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.378
69	resnet50_robust_l2_eps3	.369
70	imagenet_l2_3_0	.369
71	resnet-50-robust	.369
72	densenet-201	.363
73	mehrer2020_alexnet_training_seed_01_fov12	.360
74	custom_model_cv_18_dagger_408	.357
75	cvt_cvt-13-224-in1k_4_LucyV4	.356
76	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.356
77	deit_base_imagenet_full_seed-0	.341
78	effnetb1_272x240	.337
79	resnet50-SIN	.334
80	resnet_SIN_IN_FT_IN	.334
81	resnet50-SIN_IN_IN	.334
82	resnet50-SIN_IN	.329
83	voneresnet-50-robust	.322
84	resnet50_robust_l2_eps1	.316
85	resnet50_hvs_wzc	.311
86	resnet50_st_wzc_seed3_2_323	.307
87	cvt_cvt-21-224-in1k_4_LucyV4	.305
88	resnet50_ewconplus_wzc_seed6_2_323	.303
89	cvt_cvt-13-384-in1k_4_LucyV4	.296
90	blt_vs	.281
91	antialiased-r50	.275
92	resnet50_ewconplus_wzc_seed2_2__323	.275
93	AlexNet_SIN_fov12	.269
94	resnet50_st_wzc_seed7_2_323	.266
95	pnasnet_large_halfPrecision	.266
96	mehrer2020_alexnet_training_seed_01_fov16	.262
97	resnet152_imagenet_full	.262
98	Res2Net50_26w_4s	.260
99	deit_small_imagenet_full_seed-0	.255
100	resnet_152_v2	.250
101	resnet50_st_wzc_seed6_2_323	.249
102	resnet50_ewconplus_wzc_seed5_2__323	.248
103	resnet_50_v1	.247
104	resnet50_tutorial	.247
105	resnet50-sup	.247
106	densenet-121	.246
107	resnet_50_v2	.244
108	resnet50_ewconplus_wzc	.240
109	resnet101_imagenet_full	.239
110	cvt_cvt-21-384-in1k_4_LucyV4	.236
111	efficientnet_b1_imagenet_full	.234
112	vonegrcnn_62e_nobn	.232
113	resnet34_imagenet_full	.228
114	resnet_50_v1_spiking	.225
115	resnet50_byol_linhead	.223
116	voneresnet-50	.220
117	efficientnet_b2_imagenet_full	.211
118	vonegrcnn_47e	.208
119	grcnn	.201
120	voneresnet-50-non_stochastic	.200
121	inception_v3	.197
122	mehrer2020_alexnet_training_seed_01	.192
123	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.191
124	mehrer2020_alexnet_training_seed_05	.188
125	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.183
126	nasnet_mobile	.179
127	convnext_small_imagenet_100_seed-0	.176
128	resnet50_vicreg_linhead	.176
129	resnet50-vitoimagevidnet8	.175
130	resnet50-VITO-8deg-cc	.175
131	AlexNet_SIN	.171
132	eicircuit_resnet18_bn_20260919	.168
133	convnext_femto_ols:d1_in1k	.166
134	mehrer2020_alexnet_training_seed_02	.163
135	alexnet	.160
136	resnet50_imagenet_full	.158
137	resnet18_imagenet_full	.156
138	mehrer2020_alexnet_training_seed_04	.155
139	efficientnet_b0_imagenet_full	.150
140	resnet_50_v1_spiking_l4a	.149
141	mehrer2020_alexnet_training_seed_03	.144
142	resnet50_simsiam_linhead	.143
143	ReAlnet09	.141
144	ReAlnet07	.140
145	ReAlnet04	.136
146	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.135
147	ReAlnet10	.135
148	vgg_16	.134
149	resnet50-moclr8deg	.133
150	mobilenet_v2_1_4_224	.133
151	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.132
152	ReAlnet08	.131
153	ijepa_vith14_linhead	.130
154	resnet50_barlowtwins_linhead	.127
155	shufflenet_v2_x1_0	.124
156	mehrer2020_alexnet_training_seed_01_fov4	.108
157	resnet50_simclr_linhead_01norm	.108
158	alexnet_training_seed_09	.000
159	alexnet_training_seed_07	.000
160	mobilenet_v2_1_0_128
161	gabor_filter_entropy_4x4_v2_forcedCosine
162	gabor_filter_entropy_4x4_v2
163	dinov2_vitl14_reg_lc_featread
164	deepeisnn_ei_imagenet_resnet18
165	ReAlnet04_cornet
166	alexnet_training_seed_10
167	alexnet_training_seed_06
168	alexnet_training_seed_03
169	alexnet_training_seed_02
170	alexnet_training_seed_01
171	hmax
172	ReAlnet01_cornet
173	compact_resnet50_robust_V4
174	VOneCORnet-S
175	ReAlnet03_cornet
176	ReAlnet01_cornet_fov4

Benchmark bibtex

None

Ceiling

0.34.

Note that scores are relative to this ceiling.

Data: Kato2026Lin

Metric: error_consistency