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("MajajHong2015public.V4-reverse_pls")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	compact_resnet50_V4	.883
2	compact_resnet50_robust_V4	.882
3	compact_vgg19_V4	.867
4	compact_cornets_V4	.833
5	compact_cornets_IT	.528
6	compact_vgg19_IT	.520
7	predictive_coding_pc	.160
8	pr1_stl10_disentangled_alpha04	.150
9	gabor_filter_entropy_4x4_v2	.108
10	gabor_filter_mean_abs_4x4_v2	.105
11	gabor_filter_entropy_4x4	.103
12	gabor_filter_entropy_4x4_v2_forcedCosine	.092
13	alexnet_training_seed_03	.063
14	hmax	.061
15	convnext_large:fb_in22k_ft_in1k	.060
16	AlexNet_SIN_fov4	.059
17	AlexNet_SIN_fov16	.059
18	resnet-50-robust	.058
19	custom_model_cv_18_dagger_408	.054
20	alexnet_training_seed_07	.051
21	alexnet_training_seed_01	.051
22	alexnet_training_seed_09	.050
23	alexnet_training_seed_06	.049
24	alexnet_training_seed_02	.048
25	vit_large_patch14_clip_336:laion2b_ft_in1k	.046
26	alexnet_training_seed_10	.045
27	dinov2_vitl14_lc	.044
28	mobilenet_v2_1_4_224	.042
29	AlexNet_SIN_fov12	.040
30	ReAlnet01_cornet_fov16	.039
31	mehrer2020_alexnet_training_seed_01_fov4	.038
32	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.036
33	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.035
34	ReAlnet01_cornet_fov12	.034
35	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.033
36	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.032
37	mehrer2020_alexnet_training_seed_01_fov16	.031
38	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.031
39	ReAlnet01_cornet_fov4	.031
40	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.031
41	vit_large_patch14_clip_224:laion2b_ft_in1k	.031
42	eva02_large_in22k	.030
43	VOneCORnet-S	.030
44	vit_large_patch14_clip_224:openai_ft_in1k	.029
45	mehrer2020_alexnet_training_seed_01_fov12	.029
46	densenet-121	.029
47	AdvProp_efficientnet-b6	.028
48	ReAlnet10_cornet	.028
49	efficientnet_b0	.027
50	ReAlnet07_cornet	.027
51	ReAlnet08_cornet	.027
52	mehrer2020_alexnet_training_seed_06	.027
53	ReAlnet02_cornet	.027
54	ReAlnet03_cornet	.026
55	AdvProp_efficientnet-b2	.026
56	AT_efficientnet-b2	.026
57	mehrer2020_alexnet_training_seed_08	.026
58	ReAlnet05_cornet	.026
59	ReAlnet06_cornet	.026
60	grcnn	.026
61	mehrer2020_alexnet_training_seed_09	.026
62	mehrer2020_alexnet_training_seed_07	.026
63	ReAlnet09_cornet	.026
64	ReAlnet01_cornet	.026
65	resnet50_robust_l2_eps1	.026
66	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.026
67	ReAlnet04_cornet	.025
68	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.025
69	ViT_L_32_imagenet1k	.025
70	mehrer2020_alexnet_training_seed_04	.025
71	deit_large_imagenet_full_seed-0	.025
72	efficientnet_b2_imagenet_full	.024
73	mehrer2020_alexnet_training_seed_05	.024
74	mehrer2020_alexnet_training_seed_01	.024
75	resnet34_ecoset_full	.024
76	AdvProp_efficientnet-b7	.024
77	mehrer2020_alexnet_training_seed_03	.023
78	vit_base_patch16_clip_224:openai_ft_in1k	.023
79	AdvProp_efficientnet-b8	.023
80	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.023
81	mehrer2020_alexnet_training_seed_02	.023
82	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.023
83	AdvProp_efficientnet-b4	.022
84	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.022
85	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.022
86	cvt_cvt-21-384-in1k_4_LucyV4	.022
87	convnext_femto_ols:d1_in1k	.022
88	deit_base_imagenet_full_seed-0	.021
89	mehrer2020_alexnet_training_seed_10	.021
90	resnet50_imagenet_100_seed-0	.021
91	cvt_cvt-13-384-in1k_4_LucyV4	.021
92	shufflenet_v2_x1_0	.020
93	convnext_xxlarge_clip	.020
94	focalnet_tiny_lrf_in1k	.020
95	resnet101_ecoset_full	.019
96	deit_small_imagenet_100_seed-0	.019
97	resnet101_imagenet_full	.019
98	deit_small_imagenet_full_seed-0	.019
99	convnext_tiny_sup	.019
100	vit_large_clip336_openai	.019
101	resnet50-SIN	.019
102	resnet152_ecoset_full	.019
103	resnet50_simsiam_linhead	.019
104	resnet152_imagenet_full	.019
105	convnext_tiny:in12k_ft_in1k	.018
106	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.018
107	vit_huge_clip336_laion	.018
108	cvt_cvt-21-224-in1k_4_LucyV4	.018
109	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.017
110	cvt_cvt-13-224-in1k_4_LucyV4	.017
111	voneresnet-50-non_stochastic	.017
112	convnext_xlarge:fb_in22k_ft_in1k	.017
113	resnet50-vitoimagevidnet8	.017
114	resnet50-VITO-8deg-cc	.017
115	resnet50_ewconplus_wzc_seed2_2__323	.017
116	AlexNet_SIN	.016
117	effnetb1_272x240	.016
118	resnet50_ewconplus_wzc	.016
119	convnext_base:clip_laiona_augreg_ft_in1k_384	.016
120	vit_relpos_base_patch16_clsgap_224:sw_in1k	.016
121	convnext_large_imagenet_full_seed-0	.015
122	resnet50_st_wzc_seed3_2_323	.015
123	Res2Net50_26w_4s	.015
124	resnet50-moclr8deg	.015
125	resnet50_barlowtwins_linhead	.015
126	resnet50_imagenet_full	.015
127	dinov2_vitg14_lc	.015
128	resnet50_ewconplus_wzc_seed5_2__323	.015
129	voneresnet-50	.014
130	resnet50_vicreg_linhead	.014
131	resnet50_st_wzc_seed7_2_323	.014
132	dinov2_vitg14_reg_lc	.014
133	resnet50_ewconplus_wzc_seed6_2_323	.014
134	resnet50_ecoset_full	.014
135	dinov2_vitl14_reg_lc	.013
136	regnety_032	.013
137	resnet18_fair	.013
138	efficientnet_b1_imagenet_full	.013
139	efficientnet_b0_imagenet_full	.013
140	antialiased-r50	.013
141	resnet34_imagenet_full	.012
142	compact_resnet50_robust_IT	.012
143	resnext101_32x8d_wsl	.012
144	compact_resnet50_IT	.012
145	alexnet	.011
146	seresnext50_32x4d	.011
147	convnext_base_fb_in1k	.011
148	resnext101_32x16d_wsl	.011
149	resnet50-SIN_IN	.011
150	convnext_small_imagenet_100_seed-0	.011
151	resnet18_imagenet_full	.011
152	resnext101_32x32d_wsl	.011
153	resnet50_tutorial	.011
154	resnet_50_v1	.011
155	resnet50-sup	.011
156	resnet_50_v1_spiking	.011
157	resnet50_simclr_linhead_01norm	.011
158	resnext101_32x48d_wsl	.010
159	resnet18_ecoset_full	.010
160	resnet50-SIN_IN_IN	.010
161	resnet_SIN_IN_FT_IN	.010
162	resnet50_byol_linhead	.010
163	resnet50_st_wzc_seed6_2_323	.010
164	eicircuit_resnet18_bn_20260919	.009
165	convnext_tiny_imagenet_full_seed-0	.009
166	convnext_small_imagenet_full_seed-0	.008
167	convnext_base_imagenet_full_seed-0	.008
168	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.007
169	resnet_50_v2	.006
170	ijepa_vith14_linhead
171	dinov2_vitl14_reg_lc_featread
172	dinov2_vitl14_lc_res336
173	pnasnet_large_halfPrecision
174	convnext_xxlarge:clip_laion2b_soup_ft_in1k

Benchmark bibtex

@article{muzellec_reverse_2026,
      title = {Reverse predictivity for bidirectional comparison of neural networks and biological brains},
      volume = {8},
      issn = {2522-5839},
      url = {https://doi.org/10.1038/s42256-026-01204-0},
      doi = {10.1038/s42256-026-01204-0},
      number = {3},
      journal = {Nature Machine Intelligence},
      author = {Muzellec, Sabine and Kar, Kohitij},
      month = mar,
      year = {2026},
      pages = {474--488},
}

Ceiling

0.88.

Note that scores are relative to this ceiling.

Data: MajajHong2015public.V4

Metric: reverse_pls