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("Zerbe2026_fmri.V4-tau-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	.406
2	ijepa_vith14_linhead	.402
3	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.403
4	resnet50_vicreg_linhead	.395
5	resnet50_barlowtwins_linhead	.389
6	vit_huge_clip336_laion	.392
7	resnet50_byol_linhead	.382
8	resnet50_simsiam_linhead	.377
9	dinov2_vitg14_reg_lc	.377
10	resnet50_simclr_linhead_01norm	.369
11	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.369
12	vit_large_clip336_openai	.369
13	dinov2_vitl14_lc_res336	.368
14	vit_large_patch14_clip_224:openai_ft_in1k	.369
15	dinov2_vitg14_lc	.367
16	convnext_large:fb_in22k_ft_in1k	.367
17	cvt_cvt-21-224-in1k_4_LucyV4	.365
18	cvt_cvt-13-224-in1k_4_LucyV4	.361
19	convnext_xxlarge_clip	.359
20	cvt_cvt-13-384-in1k_4_LucyV4	.358
21	resnet50-moclr8deg	.353
22	shufflenet_v2_x1_0	.352
23	vit_large_patch14_clip_336:laion2b_ft_in1k	.352
24	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.352
25	resnet152_imagenet_full	.352
26	ReAlnet01_cornet_fov12	.347
27	eva02_large_in22k	.351
28	convnext_xlarge:fb_in22k_ft_in1k	.351
29	cvt_cvt-21-384-in1k_4_LucyV4	.350
30	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.347
31	regnety_032	.343
32	pnasnet_large_halfPrecision	.347
33	resnet101_imagenet_full	.342
34	dinov2_vitl14_reg_lc_featread	.346
35	dinov2_vitl14_reg_lc	.346
36	ReAlnet07_cornet	.341
37	resnet152_ecoset_full	.341
38	voneresnet-50	.345
39	convnext_tiny:in12k_ft_in1k	.344
40	resnet101_ecoset_full	.339
41	ReAlnet03_cornet	.343
42	ReAlnet01_cornet	.343
43	ReAlnet09_cornet	.338
44	ReAlnet04_cornet	.343
45	ReAlnet08_cornet	.338
46	resnet50-vitoimagevidnet8	.341
47	resnet50-VITO-8deg-cc	.341
48	ReAlnet02_cornet	.337
49	ReAlnet05_cornet	.337
50	ReAlnet06_cornet	.336
51	ReAlnet10_cornet	.336
52	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.340
53	efficientnet_b2_imagenet_full	.338
54	focalnet_tiny_lrf_in1k	.335
55	seresnext50_32x4d	.335
56	ReAlnet01_cornet_fov16	.331
57	resnet_50_v2	.334
58	efficientnet_b0	.334
59	deit_small_imagenet_full_seed-0	.331
60	vit_relpos_base_patch16_clsgap_224:sw_in1k	.333
61	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.333
62	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.331
63	resnet50-SIN	.330
64	resnet50_ewconplus_wzc	.330
65	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.330
66	AT_efficientnet-b2	.327
67	AdvProp_efficientnet-b2	.327
68	effnetb1_272x240	.327
69	VOneCORnet-S	.326
70	vit_base_patch16_clip_224:openai_ft_in1k	.325
71	resnext101_32x48d_wsl	.325
72	AdvProp_efficientnet-b4	.325
73	resnet34_ecoset_full	.317
74	vit_large_patch14_clip_224:laion2b_ft_in1k	.321
75	convnext_femto_ols:d1_in1k	.320
76	resnext101_32x32d_wsl	.319
77	Res2Net50_26w_4s	.317
78	AdvProp_efficientnet-b8	.314
79	alexnet	.314
80	convnext_tiny_sup	.314
81	convnext_tiny_imagenet_full_seed-0	.315
82	antialiased-r50	.314
83	resnet_SIN_IN_FT_IN	.314
84	resnet50-SIN_IN_IN	.314
85	convnext_large_imagenet_full_seed-0	.314
86	resnet_50_v1_spiking	.313
87	resnet_50_v1	.313
88	resnet50_tutorial	.313
89	resnet50-sup	.313
90	deit_large_imagenet_full_seed-0	.307
91	resnet50-SIN_IN	.310
92	efficientnet_b1_imagenet_full	.306
93	convnext_base:clip_laiona_augreg_ft_in1k_384	.310
94	resnext101_32x8d_wsl	.308
95	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.308
96	AdvProp_efficientnet-b7	.305
97	convnext_base_imagenet_full_seed-0	.306
98	resnext101_32x16d_wsl	.306
99	dinov2_vitl14_lc	.304
100	mehrer2020_alexnet_training_seed_08	.298
101	mobilenet_v2_1_4_224	.301
102	convnext_small_imagenet_full_seed-0	.301
103	mehrer2020_alexnet_training_seed_06	.297
104	mehrer2020_alexnet_training_seed_05	.296
105	mehrer2020_alexnet_training_seed_04	.295
106	efficientnet_b0_imagenet_full	.295
107	mehrer2020_alexnet_training_seed_02	.294
108	AdvProp_efficientnet-b6	.298
109	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.299
110	resnet50_imagenet_full	.298
111	mehrer2020_alexnet_training_seed_07	.294
112	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.297
113	resnet18_imagenet_full	.292
114	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.295
115	AlexNet_SIN	.295
116	mehrer2020_alexnet_training_seed_03	.290
117	grcnn	.294
118	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.294
119	mehrer2020_alexnet_training_seed_01	.289
120	resnet34_imagenet_full	.289
121	voneresnet-50-non_stochastic	.291
122	deit_base_imagenet_full_seed-0	.287
123	resnet50_ecoset_full	.286
124	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.288
125	resnet50_imagenet_100_seed-0	.285
126	eicircuit_resnet18_bn_20260919	.282
127	resnet50_robust_l2_eps1	.282
128	AlexNet_SIN_fov12	.281
129	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.278
130	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.276
131	mehrer2020_alexnet_training_seed_10	.265
132	resnet-50-robust	.268
133	densenet-121	.263
134	gabor_filter_entropy_4x4_v2	.254
135	gabor_filter_entropy_4x4_v2_forcedCosine	.253
136	ReAlnet01_cornet_fov4	.252
137	convnext_small_imagenet_100_seed-0	.249
138	ViT_L_32_imagenet1k	.245
139	mehrer2020_alexnet_training_seed_01_fov16	.240
140	mehrer2020_alexnet_training_seed_01_fov12	.237
141	deit_small_imagenet_100_seed-0	.235
142	gabor_filter_entropy_4x4	.232
143	mehrer2020_alexnet_training_seed_01_fov4	.212
144	AlexNet_SIN_fov16	.214
145	AlexNet_SIN_fov4	.184
146	gabor_filter_mean_abs_4x4_v2	.167
147	alexnet_training_seed_06	.162
148	alexnet_training_seed_02	.152
149	alexnet_training_seed_03	.152
150	alexnet_training_seed_01	.149
151	alexnet_training_seed_09	.149
152	alexnet_training_seed_07	.147
153	alexnet_training_seed_10	.144
154	compact_resnet50_robust_V4	.074
155	hmax	.070
156	mehrer2020_alexnet_training_seed_09
157	resnet18_ecoset_full

Benchmark bibtex

@inproceedings{zerbe_laion-fmri_2026,
      title = {{LAION}-{fMRI}: A densely sampled 7T-fMRI dataset providing broad coverage of natural image diversity},
      author = {Zerbe, Josefine and Roth, Johannes and Mell, Maggie Mae and Herholz, Peer and Knapen, Tomas and Hebart, Martin N.},
      year = {2026},
      booktitle = {Vision Sciences Society Annual Meeting},
  }

Ceiling

0.79.

Note that scores are relative to this ceiling.

Data: Zerbe2026_fmri.V4-tau

Metric: ridgecv