Benchmark
MIEB(Multilingual) - Compositionality
Multilingual benchmarks compare model performance across languages. Scores are percentages.
- Models measured
- 3
- Scores recorded
- 3
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 41.6%
- Mean
- 45.1%
- Median
- 5.7%
- Std dev
- 33.6%
- 10th percentile
- 46.1%
- 90th percentile
- 33.6%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Jina Embeddings v5 Omni SmallCompositionality· MTEB Multimodal Boards data docs | 46.1% |
| 2 | Jina Clip v1Compositionality· MTEB Multimodal Boards data docs | 45.1% |
| 3 | Jina Embeddings v5 Omni NanoCompositionality· MTEB Multimodal Boards data docs | 33.6% |