Benchmark
MTEB(fas, v2) - BitextMining
Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.
- Models measured
- 12
- Scores recorded
- 12
- Sources
- 1
- IQ weight
- 0.00
Score distribution
- 54.2%
- Mean
- 57.3%
- Median
- 10.1%
- Std dev
- 44.5%
- 10th percentile
- 62.3%
- 90th percentile
- 25.9%
- Min
This benchmark has a weight of 0.00 for embedding models, measured over 12 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 v3BitextMining· MTEB Leaderboard data | 65.5% |
| 2 | Qwen3 Embedding 8BBitextMining· MTEB Leaderboard data | 62.3% |
| 3 | Qwen3 Embedding 4BBitextMining· MTEB Leaderboard data | 61.3% |
| 4 | BGE M3BitextMining· MTEB Leaderboard data | 61.1% |
| 5 | Jina Embeddings v5 Text SmallBitextMining· MTEB Leaderboard data | 57.4% |
| 6 | Jina Embeddings v5 Omni SmallBitextMining· MTEB Leaderboard data | 57.3% |
| 7 | Multilingual E5 LargeBitextMining· MTEB Leaderboard data | 56.6% |
| 8 | Multilingual E5 Large InstructBitextMining· MTEB Leaderboard data | 55.4% |
| 9 | Jina Embeddings v5 Omni NanoBitextMining· MTEB Leaderboard data | 51.8% |
| 10 | Text Embedding 3 SmallBitextMining· MTEB Leaderboard data | 51.5% |
| 11 | E5 Mistral 7B InstructBitextMining· MTEB Leaderboard data | 44.5% |
| 12 | Qwen3 Embedding 0.6BBitextMining· MTEB Leaderboard data | 25.9% |