Benchmark
Agentset Reranker nDCG@10 - arguana
Rerank benchmarks measure how well models put the most relevant search results first. Scores are percentages.
- Models measured
- 6
- Scores recorded
- 6
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 41.6%
- Mean
- 42.5%
- Median
- 5.4%
- Std dev
- 37.4%
- 10th percentile
- 51.9%
- 90th percentile
- 35.5%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen3 Reranker 8Barguana· Agentset | 51.9% |
| 2 | Rerank V4.0 Proarguana· Agentset | 43.9% |
| 3 | Rerank V4.0 Fastarguana· Agentset | 42.5% |
| 4 | BGE Reranker v2 M3arguana· Agentset | 38.6% |
| 5 | Jina Reranker v2 Base Multilingualarguana· Agentset | 37.4% |
| 6 | Rerank V3.5arguana· Agentset | 35.5% |