Skip to content

Benchmark

UniGenBench++ (Chinese) - Consistency

Scores are percentages.

Models measured
29
Scores recorded
29
Sources
1
IQ weight
1.87

Score distribution

65.7%
Mean
75%
Median
27.2%
Std dev
2.3%
10th percentile
85.7%
90th percentile
0.46%
Min
0.46%90.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.87 for image-generation models, measured over 28 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)