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_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.646
2	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.645
3	cvt_cvt-13-384-in1k_4_LucyV4	.642
4	vit_large_patch14_clip_224:laion2b_ft_in1k	.642
5	resnet_SIN_IN_FT_IN	.642
6	resnet-50-robust	.639
7	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.637
8	vit_large_patch14_clip_224:openai_ft_in1k	.636
9	resnet50-SIN	.633
10	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.632
11	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.633
12	AlexNet_SIN_fov12	.634
13	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.630
14	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.621
15	resnet50_robust_l2_eps1	.619
16	voneresnet-50	.617
17	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.617
18	convnext_femto_ols:d1_in1k	.617
19	deit_base_imagenet_full_seed-0	.617
20	vit_base_patch16_clip_224:openai_ft_in1k	.615
21	resnet_50_v1	.611
22	resnet50-sup	.611
23	resnet50_tutorial	.611
24	resnet_50_v1_spiking	.611
25	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.612
26	convnext_base:clip_laiona_augreg_ft_in1k_384	.610
27	resnet50-SIN_IN_IN	.608
28	antialiased-r50	.607
29	AlexNet_SIN	.610
30	AdvProp_efficientnet-b8	.606
31	shufflenet_v2_x1_0	.607
32	AdvProp_efficientnet-b7	.604
33	efficientnet_b2_imagenet_full	.603
34	vit_large_patch14_clip_336:laion2b_ft_in1k	.603
35	resnext101_32x16d_wsl	.599
36	vit_relpos_base_patch16_clsgap_224:sw_in1k	.600
37	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.600
38	AlexNet_SIN_fov16	.599
39	resnext101_32x48d_wsl	.593
40	AdvProp_efficientnet-b6	.591
41	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.592
42	resnet50-vitoimagevidnet8	.593
43	resnet50-VITO-8deg-cc	.593
44	grcnn	.590
45	focalnet_tiny_lrf_in1k	.590
46	convnext_xlarge:fb_in22k_ft_in1k	.590
47	convnext_large:fb_in22k_ft_in1k	.589
48	resnet50-moclr8deg	.587
49	AdvProp_efficientnet-b4	.587
50	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.588
51	deit3_base_patch16_224_fb_in22k_ft_in1k	.588
52	custom_model_cv_18_dagger_408	.587
53	deit3_base_patch16_224_fb_in1k	.585
54	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.585
55	convnext_base_fb_in22k_ft_in1k	.580
56	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.580
57	cvt_cvt-21-384-in1k_4_LucyV4	.581
58	yudixie_wbh_object_class_scaling_0415_imagenet	.579
59	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.577
60	Res2Net50_26w_4s	.577
61	cvt_cvt-13-224-in1k_4_LucyV4	.576
62	convnext_tiny_sup	.575
63	deepeisnn_ei_imagenet_resnet18	.573
64	densenet-121	.572
65	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.573
66	compact_resnet50_robust_V4	.569
67	resnext101_32x8d_wsl	.567
68	resnet50_ewconplus_wzc	.568
69	deit_large_imagenet_full_seed-0	.567
70	ReAlnet03_cornet	.566
71	resnext101_32x32d_wsl	.564
72	deit_small_imagenet_full_seed-0	.566
73	cvt_cvt-21-224-in1k_4_LucyV4	.562
74	resnet50-SIN_IN	.561
75	convnext_small_imagenet_full_seed-0	.560
76	ReAlnet01_cornet	.560
77	ReAlnet04_cornet	.559
78	mobilenet_v2_1_4_224	.555
79	efficientnet_b0	.554
80	AT_efficientnet-b2	.553
81	AdvProp_efficientnet-b2	.553
82	deepeisnn_ei_cifar10_resnet18	.553
83	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.544
84	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.544
85	resnet152_imagenet_full	.541
86	convnext_base_imagenet_full_seed-0	.535
87	deepeisnn_ei_cifar10_vgg19	.526
88	resnet_50_v2	.525
89	alexnet_training_seed_02	.522
90	alexnet_training_seed_09	.521
91	alexnet_training_seed_01	.518
92	alexnet_training_seed_07	.516
93	ViT_L_32_imagenet1k	.504
94	alexnet_training_seed_10	.500
95	resnet50_imagenet_full	.499
96	alexnet_training_seed_06	.485
97	yudixie_wbh_object_class_scaling_0415_run_0	.473
98	yudixie_wbh_object_class_scaling_0415_run_2	.469
99	convnext_tiny:in12k_ft_in1k	.469
100	effnetb1_272x240	.458
101	convnext_large_imagenet_full_seed-0	.460
102	convnext_tiny_imagenet_full_seed-0	.457
103	alexnet_training_seed_03	.452
104	yudixie_wbh_object_class_scaling_0415_run_1	.442
105	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.403
106	yudixie_wbh_object_class_scaling_0415_run_3	.394
107	pr1_stl10_disentangled_alpha04	.348
108	predictive_coding_pc	.343
109	ReAlnet01_cornet_fov4	.291
110	convnext_base_fb_in1k	.257
111	convnext_base_fb_in1k_gram_aligned_dinov3	.234
112	hmax	.212
113	alexnet
114	resnet18_fair
115	resnet50_st_wzc_seed3_2_323
116	resnet50_st_wzc_seed5_2_323
117	resnet50_st_wzc_seed4_2_323
118	resnet50_st_wzc_noseed_2_323
119	resnet50_st_wzc_2_323
120	resnet50_ewconplus_wzc_seed7_2_323
121	resnet50_ewconplus_wzc_2_323
122	resnet50_ewconplus_wzc_seed5_2__323
123	resnet50_ewconplus_wzc_seed6_2_323
124	resnet50_ewconplus_wzc_seed3_2_323
125	resnet50_st_wzc_seed7_2_323
126	resnet50_ewconplus_wzc_seed4_2_323
127	resnet50_st_wzc_seed6_2_323
128	resnet50_ewconplus_wzc_seed2_2__323
129	VOneCORnet-S
130	voneresnet-50-non_stochastic
131	pnasnet_large_halfPrecision

Benchmark bibtex

None

Ceiling

0.54.

Note that scores are relative to this ceiling.

Data: Hebart2023_fmri.V1

Metric: ridgecv