Skip to content

Benchmark

MMEB Image - GQA

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

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

Score distribution

92.6%
Mean
92.6%
Median
0%
Std dev
92.6%
10th percentile
92.6%
90th percentile
92.6%
Min

Leaderboard

Data credits: MMEB Leaderboard (TIGER-Lab), Apache-2.0 (licence)