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("Hebart2023_fmri.V1-ridgecv")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	vit_large_clip336_openai	.646
2	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.646
3	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.645
4	cvt_cvt-13-384-in1k_4_LucyV4	.642
5	vit_large_patch14_clip_224:laion2b_ft_in1k	.642
6	resnet_SIN_IN_FT_IN	.642
7	resnet-50-robust	.639
8	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.637
9	vit_large_patch14_clip_224:openai_ft_in1k	.636
10	resnet50-SIN	.633
11	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.632
12	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.633
13	vit_huge_clip336_laion	.632
14	AlexNet_SIN_fov12	.634
15	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.630
16	convnext_large_in22k_supervised	.629
17	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.621
18	resnet50_robust_l2_eps1	.619
19	voneresnet-50	.617
20	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.617
21	convnext_femto_ols:d1_in1k	.617
22	deit_base_imagenet_full_seed-0	.617
23	vit_base_patch16_clip_224:openai_ft_in1k	.615
24	dinov2_vitl14_lc_res336	.614
25	resnet_50_v1	.611
26	resnet50-sup	.611
27	resnet50_tutorial	.611
28	resnet_50_v1_spiking	.611
29	dinov2_vitl14_lc	.611
30	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.612
31	convnext_xxlarge_clip	.611
32	convnext_base:clip_laiona_augreg_ft_in1k_384	.610
33	resnet50-SIN_IN_IN	.608
34	antialiased-r50	.607
35	AlexNet_SIN	.610
36	AdvProp_efficientnet-b8	.606
37	shufflenet_v2_x1_0	.607
38	AdvProp_efficientnet-b7	.604
39	efficientnet_b2_imagenet_full	.603
40	vit_large_patch14_clip_336:laion2b_ft_in1k	.603
41	dinov2_vitg14_lc	.603
42	dinov2_vitg14_reg_lc	.603
43	seresnext50_32x4d	.599
44	resnext101_32x16d_wsl	.599
45	vit_relpos_base_patch16_clsgap_224:sw_in1k	.600
46	convnext_large_clip_laion2b_finelayers	.599
47	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.600
48	AlexNet_SIN_fov16	.599
49	resnext101_32x48d_wsl	.593
50	AdvProp_efficientnet-b6	.591
51	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.592
52	resnet50-vitoimagevidnet8	.593
53	resnet50-VITO-8deg-cc	.593
54	grcnn	.590
55	dinov2_vitl14_reg_lc_featread	.590
56	dinov2_vitl14_reg_lc	.590
57	focalnet_tiny_lrf_in1k	.590
58	convnext_xlarge:fb_in22k_ft_in1k	.590
59	convnext_large:fb_in22k_ft_in1k	.589
60	resnet50-moclr8deg	.587
61	AdvProp_efficientnet-b4	.587
62	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.588
63	deit3_base_patch16_224_fb_in22k_ft_in1k	.588
64	custom_model_cv_18_dagger_408	.587
65	deit3_base_patch16_224_fb_in1k	.585
66	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.585
67	convnext_base_fb_in22k_ft_in1k	.580
68	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.580
69	cvt_cvt-21-384-in1k_4_LucyV4	.581
70	yudixie_wbh_object_class_scaling_0415_imagenet	.579
71	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.577
72	Res2Net50_26w_4s	.577
73	cvt_cvt-13-224-in1k_4_LucyV4	.576
74	convnext_tiny_sup	.575
75	deepeisnn_ei_imagenet_resnet18	.573
76	densenet-121	.572
77	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.573
78	compact_resnet50_robust_V4	.569
79	resnext101_32x8d_wsl	.567
80	resnet50_ewconplus_wzc	.568
81	deit_large_imagenet_full_seed-0	.567
82	ReAlnet03_cornet	.566
83	resnext101_32x32d_wsl	.564
84	deit_small_imagenet_full_seed-0	.566
85	cvt_cvt-21-224-in1k_4_LucyV4	.562
86	resnet50-SIN_IN	.561
87	convnext_small_imagenet_full_seed-0	.560
88	ReAlnet01_cornet	.560
89	ReAlnet04_cornet	.559
90	mobilenet_v2_1_4_224	.555
91	efficientnet_b0	.554
92	regnety_032	.555
93	AT_efficientnet-b2	.553
94	AdvProp_efficientnet-b2	.553
95	deepeisnn_ei_cifar10_resnet18	.553
96	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.544
97	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.544
98	resnet152_imagenet_full	.541
99	convnext_base_imagenet_full_seed-0	.535
100	deepeisnn_ei_cifar10_vgg19	.526
101	resnet_50_v2	.525
102	alexnet_training_seed_02	.522
103	alexnet_training_seed_09	.521
104	gabor_filter_entropy_4x4_v2_forcedCosine	.517
105	alexnet_training_seed_01	.518
106	gabor_filter_entropy_4x4_v2	.514
107	alexnet_training_seed_07	.516
108	eicircuit_resnet18_bn_20260919	.515
109	ViT_L_32_imagenet1k	.504
110	alexnet_training_seed_10	.500
111	resnet50_imagenet_full	.499
112	gabor_filter_entropy_4x4	.488
113	alexnet_training_seed_06	.485
114	yudixie_wbh_object_class_scaling_0415_run_0	.473
115	yudixie_wbh_object_class_scaling_0415_run_2	.469
116	convnext_tiny:in12k_ft_in1k	.469
117	effnetb1_272x240	.458
118	convnext_large_imagenet_full_seed-0	.460
119	convnext_tiny_imagenet_full_seed-0	.457
120	alexnet_training_seed_03	.452
121	yudixie_wbh_object_class_scaling_0415_run_1	.442
122	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.403
123	yudixie_wbh_object_class_scaling_0415_run_3	.394
124	pr1_stl10_disentangled_alpha04	.348
125	predictive_coding_pc	.343
126	ReAlnet01_cornet_fov4	.291
127	convnext_base_fb_in1k	.257
128	gabor_filter_mean_abs_4x4_v2	.258
129	convnext_base_fb_in1k_gram_aligned_dinov3	.234
130	hmax	.212
131	eva02_large_in22k
132	alexnet
133	resnet18_fair
134	resnet50_st_wzc_seed3_2_323
135	resnet50_ewconplus_wzc_seed5_2__323
136	resnet50_ewconplus_wzc_seed6_2_323
137	resnet50_st_wzc_seed7_2_323
138	resnet50_st_wzc_seed6_2_323
139	resnet50_ewconplus_wzc_seed2_2__323
140	VOneCORnet-S
141	voneresnet-50-non_stochastic
142	pnasnet_large_halfPrecision

Benchmark bibtex

None

Ceiling

0.54.

Note that scores are relative to this ceiling.

Data: Hebart2023_fmri.V1

Metric: ridgecv