Skip to content

Benchmark

AHELM: Auditory Perception - Mean win rate

Scores are percentages. Higher is better.

Models measured
12
Scores recorded
12
Sources
1
IQ weight
1.40

Score distribution

52%
Mean
51.6%
Median
21.5%
Std dev
26.6%
10th percentile
73.4%
90th percentile
14.1%
Min
14.1%93.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.40 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