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-accuracy_distance")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	convnext_xxlarge:clip_laion2b_soup_ft_in1k	1.0
2	dinov2_vitl14_reg_lc	1.0
3	dinov2_vitl14_lc_res336	1.0
4	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	1.0
5	dinov2_vitl14_lc	1.0
6	vit_huge_patch14_clip_336:laion2b_ft_in12k_in1k	1.0
7	convnext_xxlarge_clip	1.0
8	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	1.0
9	dinov2_vitg14_lc	1.0
10	vit_huge_clip336_laion	.997
11	dinov2_vitg14_reg_lc	.993
12	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.990
13	eva02_large_in22k	.990
14	vit_large_patch14_clip_224:laion2b_ft_in1k	.981
15	convnext_large_clip_laion2b_finelayers	.962
16	vit_large_patch14_clip_224:openai_ft_in1k	.945
17	vit_large_patch14_clip_336:laion2b_ft_in1k	.929
18	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.927
19	convnext_base_fb_in22k_ft_in1k	.921
20	vit_large_clip336_openai	.912
21	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.912
22	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.907
23	resnext101_32x32d_wsl	.900
24	convnext_large:fb_in22k_ft_in1k	.897
25	convnext_xlarge:fb_in22k_ft_in1k	.896
26	convnext_large_in22k_supervised	.893
27	resnext101_32x48d_wsl	.893
28	resnext101_32x16d_wsl	.886
29	AdvProp_efficientnet-b7	.851
30	AdvProp_efficientnet-b4	.843
31	convnext_base:clip_laiona_augreg_ft_in1k_384	.838
32	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.817
33	resnext101_32x8d_wsl	.811
34	AdvProp_efficientnet-b6	.808
35	deit3_base_patch16_224_fb_in22k_ft_in1k	.798
36	AdvProp_efficientnet-b8	.778
37	convnext_base_fb_in1k	.757
38	deit3_base_patch16_224_fb_in1k	.751
39	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.742
40	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.741
41	vit_base_patch16_clip_224:openai_ft_in1k	.703
42	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.682
43	regnety_032	.674
44	vit_relpos_base_patch16_clsgap_224:sw_in1k	.640
45	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.630
46	convnext_base_imagenet_full_seed-0	.624
47	cv_18_dagger_408_pretrained	.621
48	alexnet_training_seed_07	.616
49	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.610
50	AT_efficientnet-b2	.601
51	AdvProp_efficientnet-b2	.601
52	convnext_tiny:in12k_ft_in1k	.600
53	convnext_large_imagenet_full_seed-0	.599
54	focalnet_tiny_lrf_in1k	.598
55	convnext_small_imagenet_full_seed-0	.597
56	convnext_base_fb_in1k_gram_aligned_dinov3	.559
57	antialias-resnet152	.559
58	convnext_tiny_imagenet_full_seed-0	.559
59	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.550
60	effnetb1_272x240	.539
61	ViT_L_32_imagenet1k	.525
62	resnet50_robust_l2_eps3	.524
63	imagenet_l2_3_0	.524
64	resnet-50-robust	.524
65	deit_large_imagenet_full_seed-0	.524
66	seresnext50_32x4d	.522
67	convnext_tiny_sup	.517
68	resnet50-SIN	.512
69	resnet50_st_wzc_2	.501
70	voneresnet-50-robust	.495
71	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.473
72	resnet_101_v1	.470
73	efficientnet_b0	.459
74	deit_base_imagenet_full_seed-0	.458
75	cvt_cvt-13-224-in1k_4_LucyV4	.455
76	AlexNet_SIN_fov12	.452
77	grcnn_109	.449
78	AlexNet_SIN_fov16	.447
79	custom_model_cv_18_dagger_408	.447
80	resnet_101_v2	.442
81	cvt_cvt-13-384-in1k_4_LucyV4	.440
82	mehrer2020_alexnet_training_seed_01_fov12	.437
83	densenet-201	.425
84	resnet50_st_wzc_seed7_2_323	.413
85	blt_vs	.413
86	mehrer2020_alexnet_training_seed_01_fov16	.412
87	resnet50-SIN_IN	.411
88	resnet50_robust_l2_eps1	.407
89	resnet50_ewconplus_wzc_seed2_2__323	.393
90	resnet50_st_wzc_seed3_2_323	.383
91	resnet152_imagenet_full	.383
92	resnet101_imagenet_full	.383
93	resnet_SIN_IN_FT_IN	.381
94	resnet50-SIN_IN_IN	.381
95	resnet50_hvs_wzc	.381
96	densenet-121	.380
97	resnet50_ewconplus_wzc	.370
98	cvt_cvt-21-224-in1k_4_LucyV4	.370
99	cvt_cvt-21-384-in1k_4_LucyV4	.369
100	pnasnet_large_halfPrecision	.364
101	resnet50_st_wzc_seed6_2_323	.363
102	resnet50_ewconplus_wzc_seed6_2_323	.359
103	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.341
104	Res2Net50_26w_4s	.337
105	AlexNet_SIN	.333
106	inception_v3	.331
107	antialiased-r50	.328
108	efficientnet_b1_imagenet_full	.321
109	resnet50-vitoimagevidnet8	.320
110	resnet50-VITO-8deg-cc	.320
111	deit_small_imagenet_full_seed-0	.318
112	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.315
113	resnet_50_v1	.309
114	resnet50_tutorial	.309
115	resnet50-sup	.309
116	efficientnet_b2_imagenet_full	.308
117	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.307
118	resnet50_ewconplus_wzc_seed5_2__323	.305
119	mehrer2020_alexnet_training_seed_01	.304
120	vonegrcnn_62e_nobn	.302
121	resnet50_vicreg_linhead	.299
122	resnet_50_v2	.298
123	voneresnet-50-non_stochastic	.298
124	nasnet_mobile	.296
125	resnet50_byol_linhead	.294
126	vonegrcnn_47e	.294
127	resnet_50_v1_spiking	.293
128	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.293
129	mehrer2020_alexnet_training_seed_05	.292
130	resnet34_imagenet_full	.292
131	grcnn	.291
132	resnet_152_v2	.289
133	resnet50_simsiam_linhead	.289
134	mehrer2020_alexnet_training_seed_03	.288
135	resnet50-moclr8deg	.284
136	mobilenet_v2_1_4_224	.278
137	mehrer2020_alexnet_training_seed_02	.277
138	mehrer2020_alexnet_training_seed_04	.276
139	voneresnet-50	.273
140	convnext_femto_ols:d1_in1k	.263
141	efficientnet_b0_imagenet_full	.246
142	convnext_small_imagenet_100_seed-0	.244
143	eicircuit_resnet18_bn_20260919	.239
144	resnet50_imagenet_full	.236
145	alexnet_training_seed_03	.235
146	shufflenet_v2_x1_0	.214
147	alexnet	.212
148	ReAlnet09	.209
149	resnet50_simclr_linhead_01norm	.208
150	resnet50_barlowtwins_linhead	.200
151	ijepa_vith14_linhead	.199
152	resnet_50_v1_spiking_l4a	.194
153	resnet18_imagenet_full	.190
154	ReAlnet08	.189
155	mehrer2020_alexnet_training_seed_01_fov4	.188
156	ReAlnet10	.183
157	vgg_16	.182
158	ReAlnet04	.180
159	ReAlnet07	.180
160	alexnet_training_seed_09	.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_01
169	alexnet_training_seed_02
170	hmax
171	ReAlnet01_cornet
172	compact_resnet50_robust_V4
173	VOneCORnet-S
174	ReAlnet03_cornet
175	ReAlnet04_cornet
176	ReAlnet01_cornet_fov4

Benchmark bibtex

None

Ceiling

0.95.

Note that scores are relative to this ceiling.

Data: Kato2026Lin

Metric: accuracy_distance