Skip to content

Benchmark

CharXiv Reasoning Questions - NG

Vision benchmarks measure how well models understand images. Scores are percentages. Higher is better.

Models measured
21
Scores recorded
21
Sources
1
IQ weight
1.83

Score distribution

27.9%
Mean
26.6%
Median
11.1%
Std dev
17%
10th percentile
41.1%
90th percentile
8.3%
Min
8.3%60.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.83 for chat models, measured over 21 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

Data credits: CharXiv (Wang et al., NeurIPS 2024), Princeton NLP. https://charxiv.github.io