Skip to content

Benchmark

UniGenBench++ (English Long) - Style

Scores are percentages.

Models measured
32
Scores recorded
32
Sources
1
IQ weight
1.46

Score distribution

94.9%
Mean
96.5%
Median
4.4%
Std dev
89.3%
10th percentile
99.1%
90th percentile
81.8%
Min
81.8%99.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.46 for image-generation models, measured over 31 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: UniGenBench++ (CodeGoat24), MIT (licence)