Skip to content

Benchmark

MMMU - Tech & Engineering

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

Models measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

37.5%
Mean
37.7%
Median
0.2%
Std dev
37.3%
10th percentile
37.7%
90th percentile
37.3%
Min
37.3%37.7%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Pixtral 12BTech & Engineering, 2409· EvalAI data37.7%
2LLaVA 1.6 34BTech & Engineering· EvalAI data37.3%