Skip to content

Benchmark

MIEB(Img) - Any2AnyRetrieval

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

Models measured
3
Scores recorded
3
Sources
1
IQ weight
No weight published

Score distribution

32.4%
Mean
31%
Median
3.5%
Std dev
29%
10th percentile
37.3%
90th percentile
29%
Min
29%37.3%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard