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("Kato2026Tex-error_consistency")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	effnetb1_272x240	1.0
2	convnext_base_fb_in1k_gram_aligned_dinov3	1.0
3	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.983
4	Res2Net50_26w_4s	.975
5	resnet_50_v1_spiking	.965
6	resnext101_32x32d_wsl	.957
7	convnext_xxlarge_clip	.944
8	eicircuit_resnet18_bn_20260919	.942
9	cvt_cvt-13-384-in1k_4_LucyV4	.939
10	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.934
11	convnext_base_fb_in1k	.933
12	vit_large_patch14_clip_224:openai_ft_in1k	.932
13	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.928
14	dinov2_vitl14_lc	.927
15	convnext_tiny:in12k_ft_in1k	.925
16	dinov2_vitg14_reg_lc	.921
17	convnext_tiny_sup	.919
18	resnet50_byol_linhead	.911
19	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.907
20	dinov2_vitl14_reg_lc	.905
21	vit_base_patch16_clip_224:openai_ft_in1k	.902
22	resnet50-SIN_IN	.901
23	mehrer2020_alexnet_training_seed_05	.897
24	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.895
25	antialiased-r50	.894
26	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.893
27	dinov2_vitl14_lc_res336	.888
28	focalnet_tiny_lrf_in1k	.882
29	resnet_SIN_IN_FT_IN	.880
30	resnet50-SIN_IN_IN	.880
31	vit_large_clip336_openai	.878
32	convnext_base:clip_laiona_augreg_ft_in1k_384	.871
33	inception_v3	.870
34	resnet_50_v1_spiking_l4a	.866
35	resnet_50_v1	.864
36	resnet50_tutorial	.864
37	resnet50-sup	.864
38	mehrer2020_alexnet_training_seed_01	.863
39	vit_large_patch14_clip_336:laion2b_ft_in1k	.862
40	vit_large_patch14_clip_224:laion2b_ft_in1k	.860
41	mehrer2020_alexnet_training_seed_02	.855
42	cv_18_dagger_408_pretrained	.851
43	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.849
44	grcnn	.844
45	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.842
46	dinov2_vitg14_lc	.841
47	resnext101_32x8d_wsl	.840
48	vonegrcnn_47e	.838
49	resnet_101_v1	.838
50	mehrer2020_alexnet_training_seed_04	.826
51	resnet50_simclr_linhead_01norm	.822
52	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.821
53	efficientnet_b1_imagenet_full	.820
54	resnet18_imagenet_full	.819
55	eva02_large_in22k	.814
56	efficientnet_b0	.814
57	deit3_base_patch16_224_fb_in1k	.814
58	convnext_large_in22k_supervised	.813
59	resnet50_imagenet_full	.811
60	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.799
61	mehrer2020_alexnet_training_seed_03	.794
62	mobilenet_v2_1_4_224	.793
63	resnet50_ewconplus_wzc_seed6_2_323	.790
64	convnext_base_fb_in22k_ft_in1k	.789
65	resnet101_imagenet_full	.783
66	resnet50_simsiam_linhead	.783
67	cvt_cvt-21-224-in1k_4_LucyV4	.778
68	resnet50_st_wzc_2	.774
69	densenet-201	.773
70	convnext_tiny_imagenet_full_seed-0	.771
71	resnet50_ewconplus_wzc_seed2_2__323	.769
72	grcnn_109	.768
73	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.767
74	resnext101_32x48d_wsl	.764
75	vit_huge_clip336_laion	.764
76	resnet_152_v2	.763
77	antialias-resnet152	.762
78	voneresnet-50	.758
79	ReAlnet04	.757
80	ReAlnet07	.757
81	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.752
82	convnext_base_imagenet_full_seed-0	.752
83	resnet50_robust_l2_eps3	.748
84	imagenet_l2_3_0	.748
85	resnet-50-robust	.748
86	vit_huge_patch14_clip_336:laion2b_ft_in12k_in1k	.746
87	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.746
88	densenet-121	.743
89	ijepa_vith14_linhead	.741
90	resnext101_32x16d_wsl	.738
91	pnasnet_large_halfPrecision	.735
92	blt_vs	.734
93	seresnext50_32x4d	.734
94	vonegrcnn_62e_nobn	.734
95	resnet50_ewconplus_wzc	.732
96	ViT_L_32_imagenet1k	.729
97	resnet50_robust_l2_eps1	.729
98	ReAlnet09	.728
99	convnext_xlarge:fb_in22k_ft_in1k	.728
100	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.726
101	convnext_large:fb_in22k_ft_in1k	.725
102	deit_base_imagenet_full_seed-0	.725
103	resnet50_st_wzc_seed7_2_323	.721
104	AdvProp_efficientnet-b2	.721
105	AT_efficientnet-b2	.721
106	voneresnet-50-non_stochastic	.720
107	vgg_16	.716
108	deit3_base_patch16_224_fb_in22k_ft_in1k	.714
109	cvt_cvt-21-384-in1k_4_LucyV4	.712
110	vit_relpos_base_patch16_clsgap_224:sw_in1k	.708
111	resnet_101_v2	.708
112	efficientnet_b0_imagenet_full	.705
113	resnet34_imagenet_full	.703
114	convnext_small_imagenet_full_seed-0	.700
115	AdvProp_efficientnet-b4	.698
116	resnet_50_v2	.696
117	resnet50-moclr8deg	.695
118	regnety_032	.694
119	shufflenet_v2_x1_0	.694
120	cvt_cvt-13-224-in1k_4_LucyV4	.694
121	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.679
122	convnext_large_clip_laion2b_finelayers	.674
123	efficientnet_b2_imagenet_full	.667
124	resnet50-vitoimagevidnet8	.666
125	resnet50-VITO-8deg-cc	.666
126	voneresnet-50-robust	.664
127	resnet50_hvs_wzc	.664
128	deit_large_imagenet_full_seed-0	.658
129	resnet50_vicreg_linhead	.657
130	convnext_large_imagenet_full_seed-0	.648
131	deit_small_imagenet_full_seed-0	.644
132	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.640
133	convnext_femto_ols:d1_in1k	.634
134	resnet50_st_wzc_seed3_2_323	.630
135	mehrer2020_alexnet_training_seed_01_fov12	.629
136	ReAlnet10	.627
137	convnext_small_imagenet_100_seed-0	.622
138	resnet50_ewconplus_wzc_seed5_2__323	.613
139	resnet50_st_wzc_seed6_2_323	.605
140	resnet152_imagenet_full	.602
141	mehrer2020_alexnet_training_seed_01_fov4	.591
142	ReAlnet08	.570
143	resnet50_barlowtwins_linhead	.558
144	AlexNet_SIN_fov16	.552
145	alexnet	.549
146	AdvProp_efficientnet-b7	.532
147	nasnet_mobile	.521
148	AlexNet_SIN_fov12	.483
149	resnet50-SIN	.446
150	alexnet_training_seed_07	.425
151	custom_model_cv_18_dagger_408	.364
152	AdvProp_efficientnet-b8	.324
153	AlexNet_SIN	.318
154	mehrer2020_alexnet_training_seed_01_fov16	.301
155	AdvProp_efficientnet-b6	.253
156	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.198
157	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.160
158	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.087
159	alexnet_training_seed_09	.000
160	alexnet_training_seed_03	.000
161	mobilenet_v2_1_0_128
162	gabor_filter_entropy_4x4_v2_forcedCosine
163	gabor_filter_entropy_4x4_v2
164	dinov2_vitl14_reg_lc_featread
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
172	compact_resnet50_robust_V4
173	VOneCORnet-S
174	ReAlnet04_cornet
175	ReAlnet03_cornet
176	ReAlnet01_cornet_fov4

Benchmark bibtex

None

Ceiling

0.33.

Note that scores are relative to this ceiling.

Data: Kato2026Tex

Metric: error_consistency