Skip to content

Benchmark

AHELM: Audio Scenarios - Mean win rate

Scores are percentages. Higher is better.

Models measured
12
Scores recorded
12
Sources
1
IQ weight
1.57

Score distribution

50%
Mean
49.7%
Median
14.5%
Std dev
36.2%
10th percentile
63.3%
90th percentile
22%
Min
22%80.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.57 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