Meme Understanding

Evaluates a model’s ability to interpret culture-dependent, tricky, and humor-driven content that feels obvious to humans but is hard for AI.

90
openai/o4-mini
Average duration
8s
Average tokens
553
Average cost
$0.00
100
5s
408
opper_memes_meme_01
100
12s
988
opper_memes_meme_02
100
5s
373
opper_memes_meme_03
0
13s
1177
opper_memes_meme_04
100
5s
329
opper_memes_meme_05
100
7s
453
opper_memes_meme_06
100
6s
389
opper_memes_meme_07
100
7s
409
opper_memes_meme_08
100
7s
414
opper_memes_meme_09
100
7s
474
opper_memes_meme_10
100
7s
536
opper_memes_meme_11
0
9s
618
opper_memes_meme_12
100
6s
460
opper_memes_meme_13
100
6s
397
opper_memes_meme_14
100
9s
593
opper_memes_meme_15
100
9s
611
opper_memes_meme_16
100
7s
551
opper_memes_meme_17
100
8s
611
opper_memes_meme_18
100
7s
584
opper_memes_meme_19
100
11s
775
opper_memes_meme_20
100
7s
453
opper_memes_meme_21
100
25s
389
opper_memes_meme_22
100
7s
394
opper_memes_meme_23
100
9s
705
opper_memes_meme_24
0
11s
939
opper_memes_meme_25
100
5s
333
opper_memes_meme_26
100
10s
654
opper_memes_meme_27
100
6s
397
opper_memes_meme_28
100
10s
668
opper_memes_meme_29
100
7s
465
opper_memes_meme_30
100
7s
599
opper_memes_meme_31