Benchmark
MIEB(lite) - ZeroShotClassification
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
- 47.9%
- Mean
- 52.8%
- Median
- 15.9%
- Std dev
- 21.8%
- 10th percentile
- 64.6%
- 90th percentile
- 21.8%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | embeddinggemma-2ZeroShotClassification· MTEB Multimodal Boards data docs | 64.6% |
| 2 | Jina Embeddings v5 Omni SmallZeroShotClassification· MTEB Multimodal Boards data docs | 52.8% |
| 3 | Jina Clip v1ZeroShotClassification· MTEB Multimodal Boards data docs | 52.6% |
| 4 | Jina Embeddings v5 Omni NanoZeroShotClassification· MTEB Multimodal Boards data docs | 21.8% |