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("Cowley2026.V4-pls")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	ReAlnet01_cornet_fov16	.369
2	resnet50_barlowtwins_linhead	.361
3	resnet50_vicreg_linhead	.359
4	custom_model_cv_18_dagger_408	.356
5	resnet50_byol_linhead	.354
6	resnet50_simsiam_linhead	.354
7	ijepa_vith14_linhead	.353
8	ReAlnet01_cornet_fov12	.352
9	resnet50_simclr_linhead_01norm	.351
10	voneresnet-50	.334
11	alexnet	.332
12	convnext_tiny:in12k_ft_in1k	.330
13	shufflenet_v2_x1_0	.329
14	AT_efficientnet-b2	.329
15	AdvProp_efficientnet-b2	.329
16	AlexNet_SIN_fov12	.328
17	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.328
18	mehrer2020_alexnet_training_seed_06	.327
19	resnet50-SIN	.327
20	mehrer2020_alexnet_training_seed_05	.327
21	mehrer2020_alexnet_training_seed_08	.327
22	mehrer2020_alexnet_training_seed_03	.326
23	mehrer2020_alexnet_training_seed_07	.326
24	mehrer2020_alexnet_training_seed_09	.326
25	mehrer2020_alexnet_training_seed_02	.325
26	mehrer2020_alexnet_training_seed_04	.325
27	mehrer2020_alexnet_training_seed_01	.323
28	resnet50-moclr8deg	.321
29	vit_relpos_base_patch16_clsgap_224:sw_in1k	.322
30	AdvProp_efficientnet-b4	.322
31	AdvProp_efficientnet-b8	.318
32	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.317
33	cvt_cvt-13-224-in1k_4_LucyV4	.316
34	vit_large_patch14_clip_336:laion2b_ft_in1k	.315
35	ReAlnet06_cornet	.314
36	efficientnet_b0	.314
37	ReAlnet09_cornet	.314
38	AlexNet_SIN	.315
39	ReAlnet04_cornet	.312
40	cvt_cvt-21-224-in1k_4_LucyV4	.313
41	ReAlnet05_cornet	.312
42	ReAlnet07_cornet	.311
43	ReAlnet01_cornet	.311
44	ReAlnet08_cornet	.311
45	ReAlnet02_cornet	.311
46	ReAlnet03_cornet	.311
47	cvt_cvt-13-384-in1k_4_LucyV4	.311
48	ReAlnet10_cornet	.310
49	resnet152_imagenet_full	.309
50	VOneCORnet-S	.307
51	resnet152_ecoset_full	.308
52	mehrer2020_alexnet_training_seed_01_fov16	.306
53	resnet50-VITO-8deg-cc	.305
54	resnet50-vitoimagevidnet8	.305
55	resnet101_imagenet_full	.306
56	mehrer2020_alexnet_training_seed_01_fov12	.305
57	efficientnet_b2_imagenet_full	.303
58	cvt_cvt-21-384-in1k_4_LucyV4	.303
59	resnet50_ewconplus_wzc	.301
60	resnet101_ecoset_full	.301
61	focalnet_tiny_lrf_in1k	.298
62	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.297
63	mobilenet_v2_1_4_224	.296
64	vit_large_patch14_clip_224:openai_ft_in1k	.295
65	convnext_tiny_sup	.295
66	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.291
67	effnetb1_272x240	.289
68	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.288
69	deit_small_imagenet_full_seed-0	.287
70	deit_base_imagenet_full_seed-0	.284
71	resnet_SIN_IN_FT_IN	.283
72	resnet50-SIN_IN_IN	.283
73	convnext_large:fb_in22k_ft_in1k	.283
74	resnet50-SIN_IN	.283
75	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.282
76	resnet34_ecoset_full	.282
77	resnet_50_v1	.280
78	resnet_50_v1_spiking	.280
79	resnet50_tutorial	.280
80	resnet50-sup	.280
81	convnext_femto_ols:d1_in1k	.281
82	convnext_tiny_imagenet_full_seed-0	.281
83	convnext_xlarge:fb_in22k_ft_in1k	.276
84	resnet50_imagenet_full	.276
85	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.274
86	pnasnet_large_halfPrecision	.274
87	resnet-50-robust	.273
88	Res2Net50_26w_4s	.271
89	deit_large_imagenet_full_seed-0	.272
90	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.271
91	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.271
92	resnet50_ecoset_full	.270
93	efficientnet_b1_imagenet_full	.269
94	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.270
95	mehrer2020_alexnet_training_seed_10	.268
96	convnext_large_imagenet_full_seed-0	.268
97	convnext_base_imagenet_full_seed-0	.266
98	vit_large_patch14_clip_224:laion2b_ft_in1k	.265
99	convnext_small_imagenet_full_seed-0	.264
100	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.263
101	resnet_50_v2	.263
102	antialiased-r50	.262
103	resnext101_32x32d_wsl	.262
104	resnext101_32x16d_wsl	.262
105	resnext101_32x8d_wsl	.261
106	resnext101_32x48d_wsl	.261
107	vit_base_patch16_clip_224:openai_ft_in1k	.261
108	convnext_base:clip_laiona_augreg_ft_in1k_384	.259
109	resnet50_robust_l2_eps1	.258
110	voneresnet-50-non_stochastic	.256
111	AdvProp_efficientnet-b7	.256
112	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.254
113	AdvProp_efficientnet-b6	.250
114	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.249
115	resnet18_imagenet_full	.248
116	resnet18_ecoset_full	.248
117	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.246
118	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.245
119	efficientnet_b0_imagenet_full	.246
120	eicircuit_resnet18_bn_20260919	.239
121	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.237
122	resnet50_imagenet_100_seed-0	.236
123	grcnn	.231
124	resnet34_imagenet_full	.226
125	densenet-121	.220
126	ViT_L_32_imagenet1k	.217
127	AlexNet_SIN_fov16	.215
128	deit_small_imagenet_100_seed-0	.216
129	convnext_small_imagenet_100_seed-0	.190
130	ReAlnet01_cornet_fov4	.159
131	mehrer2020_alexnet_training_seed_01_fov4	.156
132	alexnet_training_seed_06	.132
133	alexnet_training_seed_02	.131
134	alexnet_training_seed_09	.127
135	AlexNet_SIN_fov4	.128
136	alexnet_training_seed_01	.121
137	alexnet_training_seed_10	.119
138	alexnet_training_seed_03	.119
139	alexnet_training_seed_07	.116
140	hmax	.090
141	compact_resnet50_robust_V4	.047

Benchmark bibtex

@article{cowley2026compact,
    title={Compact deep neural network models of the visual cortex},
    author={Cowley, Benjamin R and Stan, Patricia L and Pillow, Jonathan W and Smith, Matthew A},
    journal={Nature},
    volume={652},
    number={8111},
    pages={947--954},
    year={2026},
    publisher={Nature Publishing Group},
    doi={10.1038/s41586-026-10150-1},
    url={https://doi.org/10.1038/s41586-026-10150-1}}

Ceiling

0.80.

Note that scores are relative to this ceiling.

Data: Cowley2026.V4

Metric: pls