Skip to content

Benchmark

MIEB(eng) - 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

33.7%
Mean
32.4%
Median
3.1%
Std dev
30.7%
10th percentile
38%
90th percentile
30.7%
Min
30.7%38%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard