Benchmark
MVEB(text, video, v1) - VideoClassification
Vision benchmarks measure how well models understand images. Scores are percentages.
- Models measured
- 4
- Scores recorded
- 4
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 36.5%
- Mean
- 41.6%
- Median
- 16.4%
- Std dev
- 11.8%
- 10th percentile
- 57.5%
- 90th percentile
- 11.8%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen3-VL Embedding 8BVideoClassification· MTEB Multimodal Boards data docs | 57.5% |
| 2 | Jina Embeddings v5 Omni SmallVideoClassification· MTEB Multimodal Boards data docs | 41.6% |
| 3 | Jina Embeddings v5 Omni NanoVideoClassification· MTEB Multimodal Boards data docs | 35.1% |
| 4 | Qwen2.5-Omni 7BVideoClassification· MTEB Multimodal Boards data docs | 11.8% |