Skip to content

Benchmark

MVEB(v1) - VideoZeroshotClassification

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

37.2%
Mean
43.4%
Median
15%
Std dev
16.5%
10th percentile
51.7%
90th percentile
16.5%
Min
16.5%51.7%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Jina Embeddings v5 Omni SmallVideoZeroshotClassification· MTEB Multimodal Boards data docs51.7%
2Jina Embeddings v5 Omni NanoVideoZeroshotClassification· MTEB Multimodal Boards data docs43.4%
3Qwen2.5-Omni 7BVideoZeroshotClassification· MTEB Multimodal Boards data docs16.5%