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.IT-reverse_pls")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	compact_resnet50_robust_IT	.851
2	compact_resnet50_IT	.850
3	compact_vgg19_IT	.841
4	compact_cornets_IT	.835
5	compact_resnet50_robust_V4	.627
6	compact_resnet50_V4	.615
7	compact_vgg19_V4	.614
8	compact_cornets_V4	.613
9	pr1_stl10_disentangled_alpha04	.299
10	alexnet_training_seed_06	.272
11	alexnet_training_seed_03	.231
12	alexnet_training_seed_01	.218
13	alexnet_training_seed_07	.216
14	predictive_coding_pc	.206
15	alexnet_training_seed_10	.191
16	hmax	.172
17	alexnet_training_seed_09	.142
18	alexnet_training_seed_02	.131
19	resnet-50-robust	.125
20	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.117
21	custom_model_cv_18_dagger_408	.112
22	convnext_femto_ols:d1_in1k	.110
23	densenet-121	.093
24	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.092
25	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.090
26	efficientnet_b0	.087
27	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.087
28	mobilenet_v2_1_4_224	.086
29	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.085
30	vit_base_patch16_clip_224:openai_ft_in1k	.085
31	vit_large_patch14_clip_224:laion2b_ft_in1k	.084
32	vit_large_patch14_clip_224:openai_ft_in1k	.084
33	AdvProp_efficientnet-b2	.082
34	AT_efficientnet-b2	.082
35	shufflenet_v2_x1_0	.081
36	resnet_SIN_IN_FT_IN	.081
37	resnet18_fair	.080
38	convnext_base:clip_laiona_augreg_ft_in1k_384	.080
39	vit_large_patch14_clip_336:laion2b_ft_in1k	.080
40	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.078
41	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.076
42	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.075
43	convnext_tiny_imagenet_full_seed-0	.074
44	convnext_tiny:in12k_ft_in1k	.074
45	VOneCORnet-S	.074
46	efficientnet_b2_imagenet_full	.073
47	ViT_L_32_imagenet1k	.071
48	AdvProp_efficientnet-b4	.071
49	AdvProp_efficientnet-b6	.070
50	convnext_base_imagenet_full_seed-0	.068
51	convnext_small_imagenet_full_seed-0	.068
52	AdvProp_efficientnet-b7	.067
53	AlexNet_SIN_fov16	.066
54	AdvProp_efficientnet-b8	.066
55	convnext_large:fb_in22k_ft_in1k	.065
56	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.065
57	resnet50_robust_l2_eps1	.065
58	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.065
59	grcnn	.065
60	AlexNet_SIN	.063
61	effnetb1_272x240	.063
62	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.062
63	ReAlnet01_cornet_fov4	.061
64	ReAlnet01_cornet	.060
65	ReAlnet03_cornet	.059
66	ReAlnet04_cornet	.058
67	alexnet	.058
68	convnext_large_imagenet_full_seed-0	.057
69	convnext_tiny_sup	.054
70	cvt_cvt-21-384-in1k_4_LucyV4	.054
71	vit_relpos_base_patch16_clsgap_224:sw_in1k	.053
72	focalnet_tiny_lrf_in1k	.052
73	AlexNet_SIN_fov12	.051
74	cvt_cvt-13-384-in1k_4_LucyV4	.050
75	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.050
76	deit_small_imagenet_full_seed-0	.049
77	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.049
78	convnext_xlarge:fb_in22k_ft_in1k	.049
79	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.048
80	deit_large_imagenet_full_seed-0	.048
81	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.047
82	antialiased-r50	.045
83	cvt_cvt-13-224-in1k_4_LucyV4	.043
84	deit_base_imagenet_full_seed-0	.043
85	resnet50-sup	.042
86	resnet_50_v1	.042
87	resnet_50_v1_spiking	.042
88	vit_large_clip336_openai	.041
89	voneresnet-50-non_stochastic	.040
90	resnet50-SIN	.040
91	convnext_xxlarge_clip	.040
92	cvt_cvt-21-224-in1k_4_LucyV4	.040
93	dinov2_vitl14_lc	.039
94	convnext_base_fb_in1k	.039
95	dinov2_vitl14_lc_res336	.038
96	voneresnet-50	.037
97	dinov2_vitg14_reg_lc	.035
98	resnext101_32x8d_wsl	.035
99	dinov2_vitl14_reg_lc_featread	.033
100	dinov2_vitl14_reg_lc	.033
101	resnet152_imagenet_full	.033
102	dinov2_vitg14_lc	.031
103	resnext101_32x16d_wsl	.030
104	resnet50-SIN_IN	.030
105	resnext101_32x32d_wsl	.028
106	resnet50-vitoimagevidnet8	.028
107	resnet50-VITO-8deg-cc	.028
108	resnet50_ewconplus_wzc_seed2_2__323	.028
109	resnet50_ewconplus_wzc_seed6_2_323	.028
110	resnet50_ewconplus_wzc	.028
111	resnext101_32x48d_wsl	.027
112	resnet50_ewconplus_wzc_seed5_2__323	.027
113	resnet50-SIN_IN_IN	.025
114	resnet50_tutorial	.025
115	Res2Net50_26w_4s	.025
116	resnet50_st_wzc_seed3_2_323	.024
117	resnet50_st_wzc_seed7_2_323	.024
118	resnet50_st_wzc_seed6_2_323	.024
119	resnet50_imagenet_full	.024
120	resnet50-moclr8deg	.023
121	resnet_50_v2	.014
122	vit_huge_clip336_laion
123	eva02_large_in22k
124	pnasnet_large_halfPrecision
125	cvt_cvt-w24-384-in22k_finetuned-in1k_4

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

0.82.

Note that scores are relative to this ceiling.

Data: MajajHong2015public.IT

Metric: reverse_pls