Benchmark
MTEB(fas, v2)
Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.
- Models measured
- 5
- Scores recorded
- 5
- Sources
- 1
- IQ weight
- 0.00
Score distribution
- 62.1%
- Mean
- 64.3%
- Median
- 4%
- Std dev
- 54.9%
- 10th percentile
- 65.6%
- 90th percentile
- 54.9%
- Min
This benchmark has a weight of 0.00 for embedding models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | BGE M3· MTEB Leaderboard data | 65.6% |
| 2 | Jina Embeddings v3· MTEB Leaderboard data | 65.3% |
| 3 | Multilingual E5 Large· MTEB Leaderboard data | 64.3% |
| 4 | E5 Mistral 7B Instruct· MTEB Leaderboard data | 60.7% |
| 5 | Text Embedding 3 Small· MTEB Leaderboard data | 54.9% |