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

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	resnext101_32x48d_wsl	.348
2	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.347
3	resnext101_32x32d_wsl	.342
4	resnext101_32x16d_wsl	.336
5	vit_large_patch14_clip_336:laion2b_ft_in1k	.334
6	convnext_large:fb_in22k_ft_in1k	.328
7	deit_large_imagenet_full_seed-0	.329
8	vit_relpos_base_patch16_clsgap_224:sw_in1k	.324
9	eva02_large_in22k	.323
10	convnext_base:clip_laiona_augreg_ft_in1k_384	.323
11	resnet50-moclr8deg	.322
12	pnasnet_large_halfPrecision	.321
13	resnet50_simclr_linhead_01norm	.325
14	densenet-121	.319
15	vit_base_patch16_clip_224:openai_ft_in1k	.319
16	deit_small_imagenet_full_seed-0	.322
17	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.317
18	convnext_femto_ols:d1_in1k	.317
19	resnet_50_v2	.316
20	resnet152_imagenet_full	.316
21	convnext_xxlarge_clip	.314
22	convnext_tiny:in12k_ft_in1k	.313
23	focalnet_tiny_lrf_in1k	.313
24	ReAlnet07_cornet	.318
25	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.312
26	ReAlnet01_cornet	.312
27	ReAlnet02_cornet	.315
28	ReAlnet10_cornet	.314
29	ReAlnet04_cornet	.310
30	resnet50_byol_linhead	.314
31	ReAlnet06_cornet	.314
32	convnext_large_imagenet_full_seed-0	.309
33	ReAlnet05_cornet	.313
34	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.308
35	ReAlnet03_cornet	.308
36	ReAlnet09_cornet	.312
37	seresnext50_32x4d	.312
38	resnet_50_v1_spiking	.307
39	resnet_50_v1	.307
40	resnet50-sup	.307
41	convnext_base_imagenet_full_seed-0	.306
42	deit_base_imagenet_full_seed-0	.310
43	Res2Net50_26w_4s	.305
44	resnet152_ecoset_full	.309
45	resnet50_barlowtwins_linhead	.309
46	regnety_032	.308
47	dinov2_vitl14_lc	.303
48	antialiased-r50	.303
49	ReAlnet01_cornet_fov12	.307
50	dinov2_vitg14_reg_lc	.303
51	convnext_tiny_sup	.302
52	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.302
53	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.301
54	shufflenet_v2_x1_0	.301
55	resnet50-SIN_IN_IN	.300
56	dinov2_vitl14_lc_res336	.300
57	resnet50-SIN_IN	.300
58	resnet50_simsiam_linhead	.304
59	VOneCORnet-S	.299
60	resnet50_vicreg_linhead	.302
61	resnet101_ecoset_full	.302
62	voneresnet-50-non_stochastic	.297
63	resnet101_imagenet_full	.301
64	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.296
65	effnetb1_272x240	.295
66	resnet50_ecoset_full	.299
67	ReAlnet01_cornet_fov16	.299
68	vit_large_patch14_clip_224:openai_ft_in1k	.294
69	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.293
70	resnet50_tutorial	.293
71	custom_model_cv_18_dagger_408	.293
72	resnet50_imagenet_full	.292
73	eicircuit_resnet18_bn_20260919	.297
74	resnet_SIN_IN_FT_IN	.290
75	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.290
76	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.289
77	efficientnet_b2_imagenet_full	.288
78	resnet50_ewconplus_wzc	.288
79	resnet34_ecoset_full	.290
80	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.286
81	voneresnet-50	.286
82	efficientnet_b0	.285
83	efficientnet_b1_imagenet_full	.289
84	resnet50-VITO-8deg-cc	.283
85	resnet50-vitoimagevidnet8	.283
86	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.283
87	resnet50_robust_l2_eps1	.278
88	efficientnet_b0_imagenet_full	.282
89	vit_large_patch14_clip_224:laion2b_ft_in1k	.276
90	convnext_small_imagenet_full_seed-0	.274
91	mobilenet_v2_1_4_224	.273
92	dinov2_vitl14_reg_lc_featread	.272
93	dinov2_vitl14_reg_lc	.272
94	alexnet	.272
95	resnet18_imagenet_full	.275
96	resnet34_imagenet_full	.271
97	convnext_tiny_imagenet_full_seed-0	.262
98	ijepa_vith14_linhead	.261
99	ReAlnet01_cornet_fov4	.257
100	vit_huge_clip336_laion	.257
101	AdvProp_efficientnet-b2	.256
102	AT_efficientnet-b2	.256
103	resnet50-SIN	.255
104	cvt_cvt-13-224-in1k_4_LucyV4	.250
105	convnext_xlarge:fb_in22k_ft_in1k	.249
106	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.248
107	resnext101_32x8d_wsl	.246
108	AdvProp_efficientnet-b6	.246
109	AdvProp_efficientnet-b4	.244
110	cvt_cvt-13-384-in1k_4_LucyV4	.241
111	mehrer2020_alexnet_training_seed_02	.244
112	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.240
113	mehrer2020_alexnet_training_seed_05	.243
114	AdvProp_efficientnet-b7	.239
115	cvt_cvt-21-384-in1k_4_LucyV4	.234
116	mehrer2020_alexnet_training_seed_03	.235
117	resnet50_imagenet_100_seed-0	.232
118	vit_large_clip336_openai	.228
119	mehrer2020_alexnet_training_seed_09	.231
120	AdvProp_efficientnet-b8	.227
121	ViT_L_32_imagenet1k	.226
122	mehrer2020_alexnet_training_seed_07	.228
123	mehrer2020_alexnet_training_seed_10	.225
124	cvt_cvt-21-224-in1k_4_LucyV4	.221
125	resnet-50-robust	.221
126	mehrer2020_alexnet_training_seed_04	.224
127	mehrer2020_alexnet_training_seed_01	.224
128	mehrer2020_alexnet_training_seed_08	.224
129	dinov2_vitg14_lc	.221
130	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.216
131	grcnn	.214
132	AlexNet_SIN	.204
133	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.203
134	mehrer2020_alexnet_training_seed_01_fov4	.202
135	mehrer2020_alexnet_training_seed_01_fov12	.195
136	AlexNet_SIN_fov12	.169
137	convnext_small_imagenet_100_seed-0	.171
138	mehrer2020_alexnet_training_seed_01_fov16	.166
139	AlexNet_SIN_fov4	.166
140	AlexNet_SIN_fov16	.136
141	deit_small_imagenet_100_seed-0	.127
142	gabor_filter_mean_abs_4x4_v2	.054
143	gabor_filter_entropy_4x4	.048
144	alexnet_training_seed_03	.047
145	hmax	.046
146	alexnet_training_seed_06	.045
147	alexnet_training_seed_02	.045
148	alexnet_training_seed_01	.044
149	alexnet_training_seed_07	.044
150	gabor_filter_entropy_4x4_v2_forcedCosine	.044
151	gabor_filter_entropy_4x4_v2	.043
152	alexnet_training_seed_10	.042
153	alexnet_training_seed_09	.035
154	compact_resnet50_robust_V4	.021
155	mehrer2020_alexnet_training_seed_06
156	resnet18_ecoset_full
157	ReAlnet08_cornet

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

0.76.

Note that scores are relative to this ceiling.

Data: Zerbe2026_fmri.IT-tau

Metric: ridgecv