Skip to content

Benchmark

Dubesor Total

Scores are percentages.

Models measured
227
Scores recorded
270
Sources
1
IQ weight
1.57

Score distribution

48.3%
Mean
50.7%
Median
20.5%
Std dev
18.3%
10th percentile
73.4%
90th percentile
0.2%
Min
0.2%86.1%
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 227 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Showing 251–270 of 270 results.