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.IT-ridgecv")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.395
2	resnext101_32x48d_wsl	.384
3	resnext101_32x16d_wsl	.380
4	deit_large_imagenet_full_seed-0	.365
5	vit_large_patch14_clip_336:laion2b_ft_in1k	.354
6	deit_small_imagenet_full_seed-0	.348
7	convnext_femto_ols:d1_in1k	.344
8	resnet50-moclr8deg	.344
9	predictive_coding_pc	.337
10	focalnet_tiny_lrf_in1k	.336
11	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.334
12	convnext_base:clip_laiona_augreg_ft_in1k_384	.333
13	vit_relpos_base_patch16_clsgap_224:sw_in1k	.331
14	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.331
15	resnet_50_v2	.329
16	deit_base_imagenet_full_seed-0	.329
17	antialiased-r50	.327
18	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.326
19	densenet-121	.321
20	resnet_50_v1	.317
21	resnet_50_v1_spiking	.317
22	resnet50-sup	.317
23	shufflenet_v2_x1_0	.318
24	resnet50_st_wzc_seed6_2_323	.316
25	convnext_tiny_sup	.315
26	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.315
27	voneresnet-50-non_stochastic	.313
28	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.313
29	resnet50_st_wzc_seed3_2_323	.312
30	efficientnet_b0	.312
31	resnet50-SIN_IN_IN	.312
32	convnext_large_imagenet_full_seed-0	.310
33	resnet152_imagenet_full	.310
34	resnet50_st_wzc_seed7_2_323	.309
35	convnext_tiny:in12k_ft_in1k	.309
36	vit_base_patch16_clip_224:openai_ft_in1k	.308
37	resnet50_st_wzc_noseed_2_323	.308
38	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.307
39	ReAlnet01_cornet	.303
40	convnext_base_imagenet_full_seed-0	.303
41	resnet50_st_wzc_seed4_2_323	.302
42	ReAlnet03_cornet	.301
43	vit_large_patch14_clip_224:openai_ft_in1k	.301
44	resnet50_st_wzc_seed5_2_323	.300
45	resnet50-VITO-8deg-cc	.299
46	resnet50-vitoimagevidnet8	.299
47	resnet50_imagenet_full	.298
48	ReAlnet04_cornet	.296
49	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.296
50	resnet50-SIN_IN	.295
51	resnet50_st_wzc_2_323	.295
52	Res2Net50_26w_4s	.295
53	convnext_large:fb_in22k_ft_in1k	.292
54	resnet50_robust_l2_eps1	.292
55	voneresnet-50	.292
56	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.291
57	pnasnet_large_halfPrecision	.290
58	resnet50_ewconplus_wzc	.287
59	VOneCORnet-S	.285
60	vit_large_patch14_clip_224:laion2b_ft_in1k	.284
61	resnet50_ewconplus_wzc_2_323	.281
62	resnet50_tutorial	.281
63	resnet50_ewconplus_wzc_seed6_2_323	.280
64	resnet50_ewconplus_wzc_seed5_2__323	.279
65	resnet_SIN_IN_FT_IN	.279
66	mobilenet_v2_1_4_224	.278
67	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.276
68	custom_model_cv_18_dagger_408	.275
69	resnet50_ewconplus_wzc_seed4_2_323	.272
70	resnet50_ewconplus_wzc_seed3_2_323	.272
71	resnet50_ewconplus_wzc_seed2_2__323	.271
72	AdvProp_efficientnet-b7	.268
73	resnet50_ewconplus_wzc_seed7_2_323	.268
74	AT_efficientnet-b2	.266
75	AdvProp_efficientnet-b2	.266
76	efficientnet_b2_imagenet_full	.259
77	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.248
78	deit3_base_patch16_224_fb_in22k_ft_in1k	.248
79	AdvProp_efficientnet-b6	.246
80	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.245
81	convnext_small_imagenet_full_seed-0	.243
82	effnetb1_272x240	.243
83	convnext_tiny_imagenet_full_seed-0	.243
84	deepeisnn_ei_imagenet_resnet18	.238
85	AdvProp_efficientnet-b4	.237
86	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.237
87	resnet50-SIN	.233
88	convnext_xlarge:fb_in22k_ft_in1k	.230
89	AdvProp_efficientnet-b8	.230
90	resnext101_32x8d_wsl	.230
91	alexnet	.229
92	yudixie_wbh_object_class_scaling_0415_imagenet	.227
93	deit3_base_patch16_224_fb_in1k	.223
94	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.223
95	cvt_cvt-13-224-in1k_4_LucyV4	.222
96	ViT_L_32_imagenet1k	.221
97	ReAlnet01_cornet_fov4	.217
98	convnext_base_fb_in22k_ft_in1k	.211
99	cvt_cvt-13-384-in1k_4_LucyV4	.210
100	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.203
101	resnet-50-robust	.201
102	cvt_cvt-21-384-in1k_4_LucyV4	.197
103	grcnn	.192
104	cvt_cvt-21-224-in1k_4_LucyV4	.191
105	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.191
106	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.189
107	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.183
108	convnext_base_fb_in1k	.178
109	pr1_stl10_disentangled_alpha04	.172
110	AlexNet_SIN	.167
111	AlexNet_SIN_fov12	.121
112	AlexNet_SIN_fov16	.108
113	resnet18_fair	.105
114	deepeisnn_ei_cifar10_resnet18	.092
115	deepeisnn_ei_cifar10_vgg19	.087
116	convnext_base_fb_in1k_gram_aligned_dinov3	.081
117	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.070
118	hmax	.059
119	yudixie_wbh_object_class_scaling_0415_run_2	.031
120	yudixie_wbh_object_class_scaling_0415_run_0	.030
121	alexnet_training_seed_10	.027
122	alexnet_training_seed_01	.027
123	yudixie_wbh_object_class_scaling_0415_run_3	.025
124	alexnet_training_seed_02	.024
125	alexnet_training_seed_03	.024
126	alexnet_training_seed_06	.024
127	alexnet_training_seed_07	.022
128	alexnet_training_seed_09	.021
129	yudixie_wbh_object_class_scaling_0415_run_1	.020
130	compact_resnet50_robust_V4	.017
131	resnext101_32x32d_wsl

Benchmark bibtex

None

Ceiling

0.39.

Note that scores are relative to this ceiling.

Data: Hebart2023_fmri.IT

Metric: ridgecv