Benchmark
Agentset Reranker Recall@10 - FiQa
Rerank benchmarks measure how well models put the most relevant search results first. Scores are percentages. Higher is better.
- Models measured
- 6
- Scores recorded
- 6
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 12.1%
- Mean
- 13%
- Median
- 1.5%
- Std dev
- 11%
- 10th percentile
- 13.5%
- 90th percentile
- 9.3%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Rerank V4.0 ProFiQa· Agentset | 13.5% |
| 2 | BGE Reranker v2 M3FiQa· Agentset | 13% |
| 3 | Rerank V3.5FiQa· Agentset | 13% |
| 4 | Rerank V4.0 FastFiQa· Agentset | 13% |
| 5 | Qwen3 Reranker 8BFiQa· Agentset | 11% |
| 6 | Jina Reranker v2 Base MultilingualFiQa· Agentset | 9.3% |