Skip to content

Benchmark

ViDoRe(v3.1)

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

Models measured
8
Scores recorded
8
Sources
1
IQ weight
1.13

Score distribution

50.2%
Mean
56.6%
Median
10.7%
Std dev
44.4%
10th percentile
58.4%
90th percentile
26%
Min
26%61.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.13 for embedding models, measured over 8 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard