Skip to content

Benchmark

CharXiv Descriptive Questions - INEX

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

Models measured
19
Scores recorded
19
Sources
1
IQ weight
1.65

Score distribution

68.4%
Mean
74.1%
Median
14.6%
Std dev
50.4%
10th percentile
82.6%
90th percentile
33.3%
Min
33.3%84.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.65 for chat models, measured over 19 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