Benchmark
Agentset Embedding Recall@10 - MSMARCO
Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages. Higher is better.
- Models measured
- 15
- Scores recorded
- 15
- Sources
- 1
- IQ weight
- 0.89
Score distribution
- 21.9%
- Mean
- 22.1%
- Median
- 0.429%
- Std dev
- 21.2%
- 10th percentile
- 22.3%
- 90th percentile
- 20.8%
- Min
This benchmark has a weight of 0.89 for embedding models, measured over 15 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 4BMSMARCO· Agentset | 22.4% |
| 2 | Qwen3 Embedding 8BMSMARCO· Agentset | 22.3% |
| 3 | Text Embedding 3 LargeMSMARCO· Agentset | 22.3% |
| 4 | voyage-3.5-liteMSMARCO· Agentset | 22.3% |
| 5 | Gemini Embedding 2MSMARCO· Agentset | 22.1% |
| 6 | voyage-3-largeMSMARCO· Agentset | 22.1% |
| 7 | voyage-3.5MSMARCO· Agentset | 22.1% |
| 8 | voyage-4MSMARCO· Agentset | 22.1% |
| 9 | BGE M3MSMARCO· Agentset | 21.9% |
| 10 | Jina Embeddings v3MSMARCO· Agentset | 21.9% |
| 11 | Jina Embeddings v5 Text SmallMSMARCO· Agentset | 21.9% |
| 12 | Embed Multilingual V3.0MSMARCO· Agentset | 21.8% |
| 13 | Qwen3 Embedding 0.6BMSMARCO· Agentset | 21.5% |
| 14 | Text Embedding 3 SmallMSMARCO· Agentset | 21.2% |
| 15 | Text Embedding 004MSMARCO· Agentset | 20.8% |