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.V4-ridgecv")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	custom_model_cv_18_dagger_408	.459
2	cvt_cvt-21-224-in1k_4_LucyV4	.367
3	resnet50_st_wzc_seed7_2_323	.367
4	resnet50_st_wzc_2_323	.365
5	resnet152_imagenet_full	.363
6	resnet50_st_wzc_seed5_2_323	.362
7	cvt_cvt-21-384-in1k_4_LucyV4	.358
8	resnet50_st_wzc_seed3_2_323	.358
9	deit3_base_patch16_224_fb_in22k_ft_in1k_gram_aligned_dinov3	.355
10	deit3_base_patch16_224_fb_in22k_ft_in1k	.355
11	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.352
12	ReAlnet04_cornet	.351
13	convnext_large:fb_in22k_ft_in1k	.344
14	convnext_xlarge:fb_in22k_ft_in1k	.346
15	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.343
16	vit_relpos_base_patch16_clsgap_224:sw_in1k	.341
17	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.342
18	cvt_cvt-13-224-in1k_4_LucyV4	.342
19	ReAlnet01_cornet	.345
20	ReAlnet03_cornet	.344
21	voneresnet-50	.340
22	shufflenet_v2_x1_0	.339
23	cvt_cvt-13-384-in1k_4_LucyV4	.338
24	resnet50-moclr8deg	.337
25	convnext_tiny:in12k_ft_in1k	.331
26	resnet50_ewconplus_wzc	.329
27	AT_efficientnet-b2	.330
28	AdvProp_efficientnet-b2	.330
29	resnet50-VITO-8deg-cc	.327
30	resnet50-vitoimagevidnet8	.327
31	deit_small_imagenet_full_seed-0	.325
32	resnet50_ewconplus_wzc_seed3_2_323	.324
33	vit_large_patch14_clip_224:openai_ft_in1k	.324
34	resnet50_ewconplus_wzc_2_323	.322
35	resnet50_ewconplus_wzc_seed6_2_323	.324
36	resnet_50_v2	.321
37	vit_large_patch14_clip_336:laion2b_ft_in1k	.322
38	deepeisnn_ei_imagenet_resnet18	.322
39	resnet50_ewconplus_wzc_seed4_2_323	.319
40	focalnet_tiny_lrf_in1k	.320
41	efficientnet_b2_imagenet_full	.319
42	resnet50_ewconplus_wzc_seed7_2_323	.318
43	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.317
44	AdvProp_efficientnet-b4	.320
45	resnet50_ewconplus_wzc_seed5_2__323	.315
46	alexnet	.315
47	VOneCORnet-S	.316
48	resnet50-SIN	.316
49	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.313
50	resnet50_ewconplus_wzc_seed2_2__323	.312
51	convnext_tiny_sup	.311
52	Res2Net50_26w_4s	.309
53	mobilenet_v2_1_4_224	.307
54	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.307
55	AdvProp_efficientnet-b8	.306
56	resnet50-SIN_IN	.299
57	deit3_base_patch16_224_fb_in1k	.296
58	deit3_base_patch16_224_fb_in1k_gram_aligned_dinov3	.296
59	efficientnet_b0	.298
60	convnext_femto_ols:d1_in1k	.295
61	convnext_tiny_imagenet_full_seed-0	.293
62	resnet_SIN_IN_FT_IN	.293
63	resnet50-SIN_IN_IN	.293
64	resnext101_32x16d_wsl	.292
65	resnext101_32x48d_wsl	.292
66	resnet50_imagenet_full	.290
67	vit_base_patch16_clip_224:openai_ft_in1k	.291
68	resnext101_32x8d_wsl	.289
69	resnet50_st_wzc_seed6_2_323	.288
70	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.289
71	yudixie_wbh_object_class_scaling_0415_imagenet	.288
72	deit_base_imagenet_full_seed-0	.284
73	resnet50-sup	.283
74	resnet50_tutorial	.283
75	resnet_50_v1_spiking	.283
76	resnet_50_v1	.283
77	deit_large_imagenet_full_seed-0	.280
78	antialiased-r50	.281
79	convnext_base_imagenet_full_seed-0	.279
80	voneresnet-50-non_stochastic	.278
81	resnet50_st_wzc_noseed_2_323	.277
82	convnext_large_imagenet_full_seed-0	.276
83	resnet50_st_wzc_seed4_2_323	.276
84	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.275
85	vit_large_patch14_clip_224:laion2b_ft_in1k	.277
86	predictive_coding_pc	.270
87	convnext_small_imagenet_full_seed-0	.270
88	convnext_base:clip_laiona_augreg_ft_in1k_384	.270
89	AlexNet_SIN_fov12	.265
90	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.265
91	pr1_stl10_disentangled_alpha04	.263
92	ReAlnet01_cornet_fov4	.264
93	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.261
94	AlexNet_SIN	.261
95	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.259
96	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.260
97	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.256
98	grcnn	.254
99	resnet-50-robust	.252
100	resnet50_robust_l2_eps1	.249
101	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.247
102	densenet-121	.244
103	convnext_base_fb_in1k_gram_aligned_dinov3	.241
104	convnext_base_fb_in1k	.241
105	AdvProp_efficientnet-b7	.235
106	resnet18_fair	.229
107	AdvProp_efficientnet-b6	.225
108	convnext_base_fb_in22k_ft_in1k_gram_aligned_dinov3	.218
109	convnext_base_fb_in22k_ft_in1k	.218
110	AlexNet_SIN_fov16	.217
111	effnetb1_272x240	.217
112	deepeisnn_ei_cifar10_resnet18	.212
113	deepeisnn_ei_cifar10_vgg19	.207
114	ViT_L_32_imagenet1k	.200
115	yudixie_wbh_object_class_scaling_0415_run_2	.147
116	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.134
117	alexnet_training_seed_06	.113
118	alexnet_training_seed_02	.111
119	alexnet_training_seed_03	.108
120	yudixie_wbh_object_class_scaling_0415_run_1	.106
121	alexnet_training_seed_09	.102
122	yudixie_wbh_object_class_scaling_0415_run_0	.104
123	alexnet_training_seed_01	.101
124	alexnet_training_seed_07	.098
125	alexnet_training_seed_10	.096
126	hmax	.096
127	compact_resnet50_robust_V4	.077
128	yudixie_wbh_object_class_scaling_0415_run_3	.072
129	resnext101_32x32d_wsl
130	convnext_xxlarge:clip_laion2b_soup_ft_in1k
131	pnasnet_large_halfPrecision

Benchmark bibtex

None

Ceiling

0.41.

Note that scores are relative to this ceiling.

Data: Hebart2023_fmri.V4

Metric: ridgecv