Skip to content

Benchmark

AHELM: Bias - Mean win rate

Scores are percentages. Higher is better.

Models measured
12
Scores recorded
12
Sources
1
IQ weight
0.67

Score distribution

42.2%
Mean
43.8%
Median
24.1%
Std dev
12.5%
10th percentile
75%
90th percentile
6.3%
Min
6.3%87.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.67 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