Skip to content

Benchmark

Arena ELO Text Factuality - Industry Legal and Government

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
116
Scores recorded
132
Sources
1
IQ weight
2.45

Score distribution

1,464.6
Mean
1,463.8
Median
32.6
Std dev
1,428
10th percentile
1,507
90th percentile
1,353.2
Min
1,353.21,526.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.45 for chat models, measured over 116 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Showing 101–132 of 132 results.

Data credits: LMArena, CC BY 4.0 (licence)