Benchmark
Agentset Embedding nDCG@10 - MSMARCO
Embedding benchmarks measure how well models find related items, for example when matching a search to relevant results. Scores are percentages.
- Models measured
- 15
- Scores recorded
- 15
- Sources
- 1
- IQ weight
- 0.00
Score distribution
- 94.1%
- Mean
- 94.2%
- Median
- 0.872%
- Std dev
- 93.1%
- 10th percentile
- 95.4%
- 90th percentile
- 91.8%
- Min
This benchmark has a weight of 0.00 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 | Jina Embeddings v5 Text SmallMSMARCO· Agentset | 95.4% |
| 2 | Qwen3 Embedding 4BMSMARCO· Agentset | 95.4% |
| 3 | Text Embedding 3 LargeMSMARCO· Agentset | 94.7% |
| 4 | Text Embedding 3 SmallMSMARCO· Agentset | 94.6% |
| 5 | Jina Embeddings v3MSMARCO· Agentset | 94.4% |
| 6 | voyage-3.5MSMARCO· Agentset | 94.4% |
| 7 | voyage-3.5-liteMSMARCO· Agentset | 94.4% |
| 8 | voyage-3-largeMSMARCO· Agentset | 94.2% |
| 9 | BGE M3MSMARCO· Agentset | 94.1% |
| 10 | Embed Multilingual V3.0MSMARCO· Agentset | 94.1% |
| 11 | Gemini Embedding 2MSMARCO· Agentset | 93.9% |
| 12 | Qwen3 Embedding 8BMSMARCO· Agentset | 93.7% |
| 13 | Qwen3 Embedding 0.6BMSMARCO· Agentset | 93.3% |
| 14 | voyage-4MSMARCO· Agentset | 93.1% |
| 15 | Text Embedding 004MSMARCO· Agentset | 91.8% |