Skip to content

Benchmark

AHELM: COREBench - PEM

Scores are percentages.

Models measured
12
Scores recorded
12
Sources
1
IQ weight
0.55

Score distribution

67.3%
Mean
73.7%
Median
13.4%
Std dev
51.4%
10th percentile
79.9%
90th percentile
34.2%
Min
34.2%81.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.55 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