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.V2-ood-ridgecv")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	vit_huge_clip336_laion	.413
2	cvt_cvt-13-384-in1k_4_LucyV4	.402
3	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.400
4	deit_small_imagenet_full_seed-0	.383
5	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.394
6	resnet50-SIN_IN	.392
7	convnext_large:fb_in22k_ft_in1k	.393
8	vit_relpos_base_patch16_clsgap_224:sw_in1k	.391
9	dinov2_vitg14_lc	.391
10	vit_large_clip336_openai	.391
11	dinov2_vitl14_reg_lc_featread	.390
12	dinov2_vitl14_reg_lc	.390
13	dinov2_vitg14_reg_lc	.390
14	antialiased-r50	.388
15	resnet_50_v1_spiking	.389
16	resnet_50_v1	.389
17	resnet50-sup	.389
18	shufflenet_v2_x1_0	.385
19	deit_large_imagenet_full_seed-0	.370
20	focalnet_tiny_lrf_in1k	.385
21	resnet_SIN_IN_FT_IN	.385
22	resnet50-SIN_IN_IN	.385
23	efficientnet_b2_imagenet_full	.383
24	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.383
25	deit_base_imagenet_full_seed-0	.369
26	regnety_032	.383
27	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.381
28	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.381
29	dinov2_vitl14_lc	.380
30	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.377
31	voneresnet-50	.376
32	Res2Net50_26w_4s	.376
33	resnet50_byol_linhead	.363
34	convnext_tiny_sup	.375
35	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.375
36	resnext101_32x8d_wsl	.373
37	resnet_50_v2	.372
38	cvt_cvt-21-224-in1k_4_LucyV4	.372
39	voneresnet-50-non_stochastic	.373
40	resnet34_imagenet_full	.360
41	custom_model_cv_18_dagger_408	.370
42	AdvProp_efficientnet-b4	.370
43	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.370
44	resnext101_32x16d_wsl	.369
45	resnet50_tutorial	.370
46	ijepa_vith14_linhead	.358
47	resnet101_ecoset_full	.356
48	convnext_tiny:in12k_ft_in1k	.367
49	vit_large_patch14_clip_224:openai_ft_in1k	.367
50	resnet50_ewconplus_wzc	.367
51	resnet50-moclr8deg	.367
52	pnasnet_large_halfPrecision	.367
53	resnet50_simsiam_linhead	.354
54	resnet34_ecoset_full	.356
55	resnet50_robust_l2_eps1	.366
56	AdvProp_efficientnet-b6	.364
57	cvt_cvt-13-224-in1k_4_LucyV4	.365
58	convnext_large_imagenet_full_seed-0	.364
59	convnext_femto_ols:d1_in1k	.364
60	resnet152_imagenet_full	.362
61	resnet18_imagenet_full	.348
62	densenet-121	.361
63	effnetb1_272x240	.362
64	AdvProp_efficientnet-b8	.360
65	resnet50_imagenet_full	.362
66	ReAlnet07_cornet	.346
67	resnet50-SIN	.359
68	convnext_base:clip_laiona_augreg_ft_in1k_384	.358
69	ReAlnet08_cornet	.344
70	resnet152_ecoset_full	.346
71	seresnext50_32x4d	.357
72	ReAlnet01_cornet_fov12	.344
73	resnet50_simclr_linhead_01norm	.341
74	cvt_cvt-21-384-in1k_4_LucyV4	.356
75	ReAlnet01_cornet_fov16	.342
76	resnet50-vitoimagevidnet8	.355
77	resnet50-VITO-8deg-cc	.355
78	grcnn	.354
79	resnext101_32x48d_wsl	.353
80	ReAlnet09_cornet	.340
81	ReAlnet01_cornet	.352
82	ReAlnet04_cornet	.352
83	ReAlnet05_cornet	.339
84	gabor_filter_entropy_4x4_v2_forcedCosine	.339
85	ReAlnet06_cornet	.338
86	ReAlnet03_cornet	.351
87	resnet18_ecoset_full	.339
88	mobilenet_v2_1_4_224	.350
89	gabor_filter_entropy_4x4_v2	.336
90	ReAlnet10_cornet	.336
91	convnext_xlarge:fb_in22k_ft_in1k	.348
92	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.346
93	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.345
94	ReAlnet02_cornet	.331
95	resnet50_vicreg_linhead	.334
96	resnext101_32x32d_wsl	.344
97	vit_base_patch16_clip_224:openai_ft_in1k	.343
98	resnet-50-robust	.341
99	convnext_tiny_imagenet_full_seed-0	.342
100	convnext_base_imagenet_full_seed-0	.341
101	vit_large_patch14_clip_336:laion2b_ft_in1k	.340
102	AdvProp_efficientnet-b7	.337
103	convnext_small_imagenet_100_seed-0	.323
104	convnext_small_imagenet_full_seed-0	.334
105	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.334
106	convnext_xxlarge_clip	.332
107	eva02_large_in22k	.332
108	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.326
109	deit_small_imagenet_100_seed-0	.312
110	vit_large_patch14_clip_224:laion2b_ft_in1k	.321
111	alexnet	.320
112	mehrer2020_alexnet_training_seed_10	.305
113	mehrer2020_alexnet_training_seed_06	.305
114	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.316
115	mehrer2020_alexnet_training_seed_02	.304
116	mehrer2020_alexnet_training_seed_03	.303
117	mehrer2020_alexnet_training_seed_04	.304
118	mehrer2020_alexnet_training_seed_07	.303
119	gabor_filter_entropy_4x4	.301
120	mehrer2020_alexnet_training_seed_05	.303
121	mehrer2020_alexnet_training_seed_08	.302
122	mehrer2020_alexnet_training_seed_01	.303
123	resnet50_ecoset_full	.302
124	efficientnet_b0	.310
125	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.309
126	AT_efficientnet-b2	.307
127	AdvProp_efficientnet-b2	.307
128	mehrer2020_alexnet_training_seed_01_fov12	.293
129	resnet101_imagenet_full	.292
130	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.305
131	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.294
132	AlexNet_SIN	.288
133	AlexNet_SIN_fov12	.284
134	VOneCORnet-S	.280
135	AlexNet_SIN_fov16	.277
136	mehrer2020_alexnet_training_seed_01_fov16	.268
137	gabor_filter_mean_abs_4x4_v2	.197
138	ReAlnet01_cornet_fov4	.202
139	alexnet_training_seed_06	.186
140	hmax	.178
141	alexnet_training_seed_03	.180
142	ViT_L_32_imagenet1k	.176
143	mehrer2020_alexnet_training_seed_01_fov4	.167
144	alexnet_training_seed_02	.166
145	alexnet_training_seed_09	.163
146	alexnet_training_seed_10	.166
147	alexnet_training_seed_07	.160
148	alexnet_training_seed_01	.160
149	AlexNet_SIN_fov4	.142
150	compact_resnet50_robust_V4	.103
151	resnet50_barlowtwins_linhead
152	mehrer2020_alexnet_training_seed_09
153	resnet50_imagenet_100_seed-0
154	efficientnet_b1_imagenet_full
155	efficientnet_b0_imagenet_full
156	eicircuit_resnet18_bn_20260919
157	dinov2_vitl14_lc_res336

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.V2-ood

Metric: ridgecv