Skip to content

Benchmark

Arena ELO Text Factuality - Korean

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

Models measured
46
Scores recorded
52
Sources
1
IQ weight
2.08

Score distribution

1,408.1
Mean
1,410.2
Median
34.3
Std dev
1,367.4
10th percentile
1,454.1
90th percentile
1,325.5
Min
1,325.51,498.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.08 for chat models, measured over 46 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51Kimi K2 Thinking· LMArena data1,348.8
52MiniMax M2.7· LMArena data1,325.5

Showing 51–52 of 52 results.

Data credits: LMArena, CC BY 4.0 (licence)