Sample stimuli

How to use

from brainscore_vision import load_benchmark
benchmark = load_benchmark("Geirhos2021lowpass-error_consistency")
score = benchmark(my_model)

Benchmark API

Code examples

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	SWSL_resnext101_32x8d	.730
2	resnext101_32x32d_wsl	.695
3	vit_base_patch16_clip_224:openai_ft_in12k_in1k	.690
4	convnext_large:fb_in22k_ft_in1k	.667
5	vit_large_patch14_clip_224:openai_ft_in1k	.663
6	vit_huge_patch14_clip_224:laion2b_ft_in12k_in1k	.648
7	fixres_resnext101_32x48d_wsl	.630
8	resnet50-VITO-8deg-cc	.625
9	vit_large_patch14_clip_336:laion2b_ft_in1k	.623
10	SWSL_resnet50	.613
11	vit_large_patch14_clip_224:laion2b_ft_in1k	.603
12	cvt_cvt-w24-384-in22k_finetuned-in1k_4	.602
13	convnext_xxlarge:clip_laion2b_soup_ft_in1k	.600
14	vit_large_patch14_clip_224:openai_ft_in12k_in1k	.589
15	resnext101_32x48d_wsl	.579
16	vit_large_patch14_clip_224:laion2b_ft_in12k_in1k	.578
17	resnext101_32x8d_wsl	.563
18	convnext_large_mlp:clip_laion2b_augreg_ft_in1k_384	.554
19	CLIP_resnet50	.544
20	CLIP_resnet50_float32	.544
21	deit_large_imagenet_full_seed-0	.543
22	vit_large_patch14_clip_336:openai_ft_in12k_in1k	.537
23	resnext101_32x16d_wsl	.535
24	vit_base_patch16_clip_224:openai_ft_in1k	.535
25	convnext_xlarge:fb_in22k_ft_in1k	.527
26	effnetv2m_custom384	.527
27	convnext_base:clip_laiona_augreg_ft_in1k_384	.516
28	vision_transformer_vit_large_patch16_224	.499
29	resnet_152_v2	.498
30	effnetb1_272x240	.497
31	CLIP_ViT-B_32	.496
32	cvt_cvt-13-384-in1k_4_LucyV4	.492
33	convnext_large_imagenet_full_seed-0	.471
34	vit_huge_patch14_clip_336:laion2b_ft_in12k_in1k	.463
35	vit_relpos_base_patch16_clsgap_224:sw_in1k	.463
36	deit_small_imagenet_full_seed-0	.460
37	deit_base_imagenet_full_seed-0	.460
38	cvt_cvt-21-384-in22k_finetuned-in1k_4_LucyV4	.454
39	focalnet_tiny_lrf_in1k	.436
40	convnext_tiny:in12k_ft_in1k	.435
41	CLIP-RN50	.419
42	convnext_tiny_imagenet_full_seed-0	.419
43	resnet_50_v2	.415
44	pnasnet_large	.410
45	ViT_B_16_imagenet1k	.409
46	ViT_L_32_imagenet1k	.406
47	inception_v4	.400
48	resnet_101_v2	.397
49	nasnet_large	.396
50	cvt_cvt-21-384-in1k_4_LucyV4	.391
51	cvt_cvt-13-384-in22k_finetuned-in1k_4_LucyV4	.385
52	vit_relpos_base_patch32_plus_rpn_256:sw_in1k	.382
53	convnext_small_imagenet_full_seed-0	.379
54	convnext_femto_ols:d1_in1k	.379
55	resnet-152_v2_pytorch	.371
56	cvt_cvt-21-224-in1k_4_LucyV4	.368
57	cvt_cvt-13-224-in1k_4_LucyV4	.361
58	convnext_base_imagenet_full_seed-0	.354
59	effnetb1_cutmixrespatch_SAM_robust32_e2_manylayers_324x288	.352
60	swin_small_patch4_window7_224:ms_in22k_ft_in1k	.338
61	pnasnet_large-tf	.338
62	antialias-resnet152	.326
63	xception-keras	.317
64	cv_18_dagger_408_pretrained	.314
65	resnet101	.313
66	effnetb1_cutmixpatch_SAM_robust32_avge6e8e9e10_manylayers_324x288	.311
67	effnetb1_cutmixrespatch_SAM_robust32_e10_manylayers_324x288	.306
68	nasnet_large-tf	.305
69	inception_v3-tf	.299
70	inception_v4-tf	.295
71	resnet152_imagenet_full	.290
72	resnet-152_v2-tf	.285
73	resnet50_byol	.274
74	resnet_152_v1	.270
75	resnet_101_v1	.268
76	antialiased-rnext101_32x8d	.268
77	inception_v3	.260
78	convnext_tiny_sup	.260
79	AT_efficientnet-b4	.258
80	AdvProp_efficientnet-b4	.258
81	grcnn_109	.252
82	inception_v3_pytorch	.252
83	resnet50	.251
84	ViT_B_32_imagenet1k	.244
85	effnetb1_cutmixpatch_augmix_robust32_avge4e7_manylayers_324x288	.243
86	resnet50_moco_v2	.232
87	effnetb1_cutmixpatch_augmix_robust32_e5_324x288	.225
88	AdvProp_efficientnet-b6	.221
89	resnet50_finetune_cutmix_e3_robust_linf8255_e0_247x234	.218
90	densenet-121-keras	.214
91	antialiased-r50	.214
92	resnet50_imagenet_full	.213
93	voneresnet-50-non_stochastic	.209
94	convnext_small_imagenet_100_seed-0	.207
95	densenet-169-keras	.203
96	AdvProp_efficientnet-b7	.203
97	resnet34_imagenet_full	.203
98	resnet-50_v2-tf	.202
99	densenet-121	.201
100	densenet-201-keras	.197
101	grcnn	.194
102	resnet_50_v1_spiking	.190
103	resnet50-moclr8deg	.187
104	densenet-169	.186
105	mobilenet_v2_1_0_224	.186
106	resnet_50_v1	.185
107	resnet50_tutorial	.185
108	resnet50-sup	.185
109	resnet-50-pytorch	.185
110	efficientnet-b7	.185
111	vgg_19	.184
112	vit_tiny_r_s16_p8_384:augreg_in21k_ft_in1k	.180
113	BiT-S-R152x4	.175
114	pnasnet_large_pytorch	.175
115	Res2Net50_26w_4s	.174
116	xception	.168
117	resnet-50_v1-tf	.167
118	resnet101_imagenet_full	.165
119	AdvProp_efficientnet-b2	.165
120	AT_efficientnet-b2	.165
121	resnet18_imagenet_full	.163
122	densenet-201	.161
123	densenet_201_pytorch	.161
124	ReAlnet04	.156
125	custom_model_cv_18_dagger_408	.156
126	resnet-101_v1-tf	.155
127	vonegrcnn_62e_nobn	.153
128	efficientnet_b1_imagenet_full	.153
129	efficientnet_b0	.152
130	inception_v1-tf	.144
131	deit_small_imagenet_100_seed-0	.143
132	vonegrcnn_47e	.141
133	efficientnet_b2_imagenet_full	.140
134	resnet50_random_linf8_perturb	.140
135	vonegrcnn_52e_full	.139
136	resnet-152_v1-tf	.138
137	mobilenet_v2_1.0_224-tf	.135
138	ReAlnet03	.135
139	nasnet_mobile	.133
140	mobilenet_v2_1.4_224-tf	.133
141	voneresnet-50	.132
142	ReAlnet09	.132
143	vonegrcnn_62e_nobn	.130
144	resnet50-SIN_IN	.129
145	BiT-S-R50x3	.129
146	blt_vs	.129
147	mobilenet_v2_1_0_192	.127
148	BiT-S-R101x3	.127
149	ReAlnet07	.127
150	vonegrcnn_52e_full	.126
151	vonegrcnn_47e	.126
152	ReAlnet08	.126
153	resnet50_random_l2_perturb	.126
154	effnetb1_cutmix_augmix_sam_e1_5avg_424x377	.125
155	inception_v2-tf	.124
156	cornet_s	.124
157	AdvProp_efficientnet-b8	.123
158	mobilenet_v2_1_4_224	.122
159	mobilenet_v2_1-4_224_pytorch	.122
160	ReAlnet02	.121
161	voneresnet-50-robust	.119
162	mobilenet_v2_0_5_192	.116
163	ReAlnet06	.115
164	nasnet_mobile-tf	.113
165	resnet50-vitoimagevidnet8	.113
166	mobilenet_v1_0.75_224	.112
167	ReAlnet10	.112
168	mobilenet_v2_0_5_224	.111
169	vgg-19-keras	.111
170	resnet50_l2_3_robust	.110
171	resnet50_linf_4_robust	.110
172	inception_v1	.109
173	resnet50-SIN_IN_IN	.109
174	resnet_SIN_IN_FT_IN	.109
175	mobilenet_v2_1.0_160-tf	.109
176	resnet_50_v1_spiking_l4a	.108
177	mobilenet_v2_1_0_160	.108
178	resnet50_linf_8_robust	.107
179	imagenet_l2_3_0	.103
180	resnet-50-robust	.103
181	resnet50_robust_l2_eps3	.103
182	vgg_16	.100
183	ReAlnet01	.100
184	BiT-S-R50x1	.100
185	CORnet-S	.097
186	vgg-16-keras	.097
187	efficientnet_b0_imagenet_full	.096
188	mobilenet_v2_1_3_224	.096
189	mobilenet_v2_0.5_192-tf	.096
190	mobilenet_v2_0.75_192-tf	.095
191	mobilenet_v1_1.0_160	.095
192	ReAlnet05	.094
193	mobilenet_v2_1.3_224-tf	.091
194	regnet_y_400mf	.087
195	imagenet_l2_10_0	.086
196	mobilenet_v2_1.0_192-tf	.085
197	deit_small_imagenet_1_seed-0	.085
198	deit_small_imagenet_10_seed-0	.085
199	resnet50_imagenet_10_seed-0	.085
200	mobilenet_v2_0.75_160-tf	.085
201	mobilenet_v2_0.5_224-tf	.084
202	mobilenet_v2_0_75_160	.084
203	resnet50_robust_l2_eps1	.082
204	mobilenet_v1_0.75_192	.081
205	alexnet_linf_8_robust	.081
206	resnet-18	.081
207	mobilenet_v2_0.75_96	.081
208	resnet-34	.080
209	alexnet	.079
210	convnext_small_imagenet_10_seed-0	.078
211	mobilenet_v2_1.0_96	.078
212	mobilenet_v2_0_75_192	.077
213	mobilenet_v2_0.35_224	.077
214	mobilenet_v2_0.5_96	.077
215	tf_efficientnetv2_s_in21ft1k_robust_linf12255_400x400	.075
216	alexnet_training_seed_01	.074
217	alexnet_training_seed_05	.074
218	resnet50-meshes-lt-100-original-scratch	.074
219	mobilenet_v2_0.75_128	.072
220	texture_shape_resnet50_trained_on_SIN	.071
221	alexnet_random_l2_3_perturb	.070
222	alexnet_l2_3_robust	.070
223	mobilenet_v2_0.75_224-tf	.069
224	resnet-50_untrained	.067
225	vonealexnet_gaussian_noise_std4_fixed	.066
226	mobilenet_v2_1_0_128	.065
227	resnet50-SIN	.064
228	alexnet_reduced_aliasing_early_checkpoint	.063
229	mobilenet_v2_0.35_96	.061
230	mobilenet_v2_0_75_224	.061
231	mobilenet_v2_0.35_192	.061
232	alexnet_robust_correct	.060
233	shufflenet_v2_x1_0	.060
234	BiT-S-R152x2	.059
235	texture_shape_alexnet_trained_on_SIN	.059
236	alexnet_training_seed_03	.058
237	alexnet_training_seed_09	.057
238	squeezenet1_1	.057
239	mobilenet_v1_0.75_128	.056
240	alexnet_ks_torevert	.056
241	alexnet	.056
242	alexnet-baseline	.056
243	alexnet	.056
244	alexnet	.056
245	alexnet	.056
246	alexnet	.056
247	alexnet	.056
248	alexnet	.056
249	alexnet	.056
250	alexnet	.056
251	AdvProp_efficientnet-b0	.055
252	mobilenet_v1_0.5_224	.055
253	resnet50-simclr-vissl	.054
254	AlexNet_SIN	.054
255	barlow-twins-resnet50	.052
256	alexnet_random_linf8_perturb	.049
257	konkle_alexnetgn_ipcl_ref12_supervised_ipcl_aug	.049
258	resnet50_simclr	.047
259	mobilenet_v1_1.0_224	.045
260	mobilenet_v1_1.0_192	.045
261	mobilenet_v2_0.35_128	.044
262	BiT-S-R101x1	.044
263	alexnet_early_checkpoint	.043
264	resnet50-vicreg	.038
265	resnet50-vicregl0p9	.038
266	mobilenet_v2_1.0_128-tf	.037
267	artResNet18_1	.032
268	resnet_50_v1_spiking_l4	.032
269	alexnet_training_seed_04	.031
270	alexnet_training_seed_02	.031
271	alexnet_training_seed_08	.031
272	alexnet_training_seed_10	.031
273	mobilenet_v2_0.5_160	.030
274	mobilenet_v2_0.35_160	.029
275	0.5x_resnet-18_LC_w_sh_1_iter	.028
276	mobilenet_v1_0.25_160	.026
277	resnet50_FractalDB	.026
278	resnet50_imagenet_1_seed-0	.025
279	resnet_50_v1_spiking_l3	.024
280	alexnet_training_seed_06	.024
281	mobilenet_v1_0.25_224	.022
282	mobilenet_v2_0.5_128	.020
283	mobilenet_v1_0.25_192	.019
284	mobilenet_v1_0.25_128	.017
285	resnet50-meshes-lt-100-original-pretrained	.016
286	mobilenet_v1_1.0_128	.016
287	squeezenet1_0	.014
288	mobilenet_v1_0.5_192	.013
289	alexnet_training_seed_07	.013
290	mobilevit_small	.013
291	mobilenet_v1_0.75_160	.013
292	konkle_alexnetgn_ipcl_ref01_primary_model	.013
293	resnet50_imagenet_100_seed-0	.007
294	mobilenet_v1_0.5_160	.006
295	resnet50-barlow	.006
296	convnext_small_imagenet_1_seed-0	.005
297	mobilenet_v1_0.5_128	.003
298	resnet50-vicregl0p75	.003
299	yudixie_resnet50_translation_rotation_0_240908
300	0.5x_resnet-18_LC
301	CORnetZ_CIFAR10
302	resnet18-local_aggregation
303	resnet18-deepcluster
304	r3m_resnet18
305	CORnet-Z
306	resnet18-instance_recognition
307	vggface
308	my-model
309	0.5x_resnet-18_LC_w_sh_10_iter
310	r3m_resnet50
311	resnet18-colorization
312	r3m_resnet34
313	resnet-18-LC_m
314	r3m_resnet50_nocrop
315	resnet18-supervised
316	resnet-18-LC_1st_conv_m
317	resnet-18-LC_conv_init_m
318	resnet18-contrastive_multiview
319	0.5x_resnet-18
320	RN50
321	0.5x_resnet-18_LC_w_sh_100_iter
322	RN50
323	resnet18-relative_position
324	resnet18-autoencoder
325	resnet18-simclr
326	FrankRobWobv0
327	ViT_L_32
328	ViT_B_32
329	resnet18-depth_prediction
330	ViT_B_16
331	r3m_resnet34_nocrop
332	resnet18-contrastive_predictive
333	omnivore_swinT
334	resnet-18-LC_untrained
335	resnet-18_untrained
336	resnet-50x2_untrained
337	dcgan
338	unet_entire
339	v1-pyr-nodown
340	CORnetZ_CIFAR10_bs32_20_04
341	pixels
342	dorinet_cornet_z
343	my-model
344	resnet50
345	openclip
346	resnet-18-LC_w_sh_1_iter_m
347	resnet-18-LC_w_sh_10_iter_conv_init_m
348	resnet-18-LC_w_sh_100_iter_m
349	resnet50-cifar
350	resnet50_pretrained_with_retinal_waves
351	resnet50_ImageNet
352	fulltest_microblockvf_nobottleneck_freeat_nopretrain_eps2_m4
353	yudixie_resnet50_random_0_240908
354	yudixie_resnet50_imagenet1kpret_0_240908
355	yudixie_resnet50_cat_obj_class_all_latents_0_240908
356	yudixie_resnet50_object_class_0_240908
357	yudixie_resnet50_imagenet1kpret_0_240312
358	resnet-18-LC_w_sh_10_iter_m
359	yudixie_resnet50_category_class_0_240908
360	resnet18_imagenet21kP
361	yudixie_resnet50_distance_reg_0_240908
362	ConvLSTM
363	PredRNN
364	SimVP
365	briaai_rmbg_1_4
366	TAU
367	MIM
368	hmax
369	resnet-18-LC_w_sh_1_iter_conv_init_m
370	ReAlnet01_cornet
371	yolos_tiny
372	resnet152_ecoset_full
373	BiT-M-R152x2
374	yudixie_resnet18_translation_reg_0_240719
375	yudixie_resnet18_distance_reg_0_240719
376	ReAlnet03_cornet
377	yudixie_resnet18_rotation_reg_0_240719
378	ReAlnet02_cornet
379	yudixie_resnet18_imagenet1kpret_0_240719
380	resnet18_ecoset_full
381	resnet34_ecoset_full
382	resnet50_ecoset_full
383	resnet101_ecoset_full
384	BiT-M-R50x1
385	yudixie_resnet18_distance_translation_0_240719
386	yudixie_resnet50_distance_translation_0_240908
387	ReAlnet04_cornet
388	omnivore_swinB
389	yudixie_resnet18_translation_rotation_0_240719
390	yudixie_resnet18_distance_rotation_0_240719
391	yudixie_resnet50_distance_rotation_0_240908
392	yudixie_resnet18_category_class_0_240719
393	yudixie_resnet18_random_0_240719
394	ReAlnet05_cornet
395	yudixie_resnet18_cat_obj_class_all_latents_0_240719
396	yudixie_resnet50_rotation_reg_0_240908
397	BiT-M-R50x3
398	resnet-18_test_m
399	yudixie_resnet18_distance_translation_rotation_0_240719
400	yudixie_resnet18_object_class_0_240719
401	ReAlnet06_cornet
402	yudixie_resnet50_translation_reg_0_240908
403	ReAlnet07_cornet
404	ReAlnet08_cornet
405	VOneCORnet-S
406	ReAlnet09_cornet
407	ViT-B/32
408	omnivore_swinS
409	ReAlnet10_cornet
410	yudixie_resnet50_distance_translation_rotation_0_240908
411	resnet50_primary_visual_cortex
412	resnet-18-LC_w_sh_100_iter_conv_init_m

Benchmark bibtex

@article{geirhos2021partial,
              title={Partial success in closing the gap between human and machine vision},
              author={Geirhos, Robert and Narayanappa, Kantharaju and Mitzkus, Benjamin and Thieringer, Tizian and Bethge, Matthias and Wichmann, Felix A and Brendel, Wieland},
              journal={Advances in Neural Information Processing Systems},
              volume={34},
              year={2021},
              url={https://openreview.net/forum?id=QkljT4mrfs}
        }

Ceiling

0.47.

Note that scores are relative to this ceiling.

Data: Geirhos2021lowpass

Metric: error_consistency