Skip to content

Benchmark

MMEB VisDoc - ViDoRe_docvqa

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

54%
Mean
54%
Median
0%
Std dev
54%
10th percentile
54%
90th percentile
54%
Min

Leaderboard

#ModelScore
1Qwen3-VL Embedding 8BViDoRe_docvqa· MMEB (TIGER-Lab) data54%

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