Benchmark
MTEB(fas, v2) - Classification
Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.
- Models measured
- 8
- Scores recorded
- 8
- Sources
- 1
- IQ weight
- 0.00
Score distribution
- 62.9%
- Mean
- 64.7%
- Median
- 3%
- Std dev
- 59.6%
- 10th percentile
- 65.7%
- 90th percentile
- 57.6%
- Min
This benchmark has a weight of 0.00 for embedding models, measured over 8 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 | Jina Embeddings v3Classification· MTEB Leaderboard data | 66.5% |
| 2 | Multilingual E5 LargeClassification· MTEB Leaderboard data | 65.7% |
| 3 | BGE M3Classification· MTEB Leaderboard data | 65.3% |
| 4 | Qwen3 Embedding 8BClassification· MTEB Leaderboard data | 64.7% |
| 5 | Qwen3 Embedding 4BClassification· MTEB Leaderboard data | 61.8% |
| 6 | Text Embedding 3 SmallClassification· MTEB Leaderboard data | 61.8% |
| 7 | Qwen3 Embedding 0.6BClassification· MTEB Leaderboard data | 59.6% |
| 8 | E5 Mistral 7B InstructClassification· MTEB Leaderboard data | 57.6% |