Skip to content

Benchmark

AHELM: Fairness - Mean win rate

Scores are percentages. Higher is better.

Models measured
12
Scores recorded
12
Sources
1
IQ weight
0.82

Score distribution

44.8%
Mean
53.1%
Median
26.4%
Std dev
18.8%
10th percentile
75%
90th percentile
3.1%
Min
3.1%90.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.82 for chat models, measured over 9 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard