Skip to content

Benchmark

UniGenBench++ (English) - Consistency

Scores are percentages.

Models measured
32
Scores recorded
32
Sources
1
IQ weight
1.02

Score distribution

71.5%
Mean
71.3%
Median
10.9%
Std dev
58.3%
10th percentile
85.7%
90th percentile
41.7%
Min
41.7%93.1%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.02 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)