Skip to content

Benchmark

GenExam Relaxed - Overall

Scores are percentages.

Models measured
27
Scores recorded
29
Sources
1
IQ weight
0.59

Score distribution

54.6%
Mean
53.4%
Median
30.5%
Std dev
17%
10th percentile
93.7%
90th percentile
11.9%
Min
11.9%95.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.59 for image-generation models, measured over 26 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Data credits: GenExam (OpenGVLab), MIT (licence)