Skip to content

Benchmark

Design Arena Elo (models/uicomponent)

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

Models measured
141
Scores recorded
153
Sources
1
IQ weight
1.76

Score distribution

1,148.2
Mean
1,156
Median
140.8
Std dev
945
10th percentile
1,325
90th percentile
752
Min
7521,389
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.76 for chat models, measured over 141 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Showing 151–153 of 153 results.