Benchmark
MTEB(cmn, v1) - STS
Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.
- Models measured
- 13
- Scores recorded
- 13
- Sources
- 1
- IQ weight
- 1.33
Score distribution
- 52%
- Mean
- 54.5%
- Median
- 6.6%
- Std dev
- 41.7%
- 10th percentile
- 61.3%
- 90th percentile
- 40.6%
- Min
This benchmark has a weight of 1.33 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 | Qwen3 Embedding 8BSTS· MTEB Leaderboard data | 63.5% |
| 2 | Qwen3 Embedding 4BSTS· MTEB Leaderboard data | 61.3% |
| 3 | BGE multilingual gemma2STS· MTEB Leaderboard data | 55.2% |
| 4 | Jina Embeddings v5 Omni NanoSTS· MTEB Leaderboard data | 54.9% |
| 5 | Jina Embeddings v5 Omni SmallSTS· MTEB Leaderboard data | 54.8% |
| 6 | Jina Embeddings v5 Text SmallSTS· MTEB Leaderboard data | 54.8% |
| 7 | Qwen3 Embedding 0.6BSTS· MTEB Leaderboard data | 54.5% |
| 8 | bge-large-zhSTS· MTEB Leaderboard data | 53.2% |
| 9 | E5 Mistral 7B InstructSTS· MTEB Leaderboard data | 48.3% |
| 10 | Multilingual E5 Large InstructSTS· MTEB Leaderboard data | 47.9% |
| 11 | Multilingual E5 LargeSTS· MTEB Leaderboard data | 45.5% |
| 12 | Text2vec Base ChineseSTS· MTEB Leaderboard data | 41.7% |
| 13 | Text Embedding Ada 002STS· MTEB Leaderboard data | 40.6% |