Benchmark
CoREB(v1) - Retrieval
Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.
- Models measured
- 3
- Scores recorded
- 3
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 60.2%
- Mean
- 60.6%
- Median
- 3.1%
- Std dev
- 56.2%
- 10th percentile
- 63.8%
- 90th percentile
- 56.2%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen3 Embedding 8BRetrieval· MTEB Extra Text Boards data | 63.8% |
| 2 | Qwen3 Embedding 4BRetrieval· MTEB Extra Text Boards data | 60.6% |
| 3 | Qwen3 Embedding 0.6BRetrieval· MTEB Extra Text Boards data | 56.2% |