Agentset Reranker Leaderboard
Rerankers reorder retrieved documents across six datasets. A fixed embedding model retrieves 50 candidates, each reranker selects 15, and GPT-5 pairwise judgments produce Elo ratings.
This evaluation reuses retrieval datasets with a distinct pairwise scoring protocol. Results depend on the initial retriever, judge, and candidate pool. The release month follows the repository commit adding the evaluation framework.
- Made by
- Agentset
- Released
- November 2025
- Kind
- Questions that need search
- Level
- University
- Scored
- Graded by a model
- Answers
- Short answer
- Tasks
- All public
- Measures
- Preference
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- Last model added
- May 2025
- Scores refreshed
- November 18th, 2026
Zerank 2 tops the leaderboard with 1,638 Qwen3-Reranker-8B leads with 1,473
Best score per entry from the benchmark's own leaderboard, read on November 18th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen3-Reranker-8B, at 1,473. This view shows only models we track, each at its best.
- 1,638 Zerank 2
- 1,629 Cohere Rerank 4 Pro
- 1,573 Zerank 1
- 1,544 Voyage AI Rerank 2.5
- 1,539 Zerank 1 Small
- 1,520 Voyage AI Rerank 2.5 Lite
- 1,510 Cohere Rerank 4 Fast
- 1,473
Qwen3-Reranker-8B
- 1,469 Contextual AI Rerank v2 Instruct
- 1,451 Cohere Rerank 3.5
- 1,327 BAAI/BGE Reranker v2 M3
- 1,327 Jina Reranker v2 Base Multilingual
Bars start at 1,300.
- 1 Zerank 2 1,638
- 2 Cohere Rerank 4 Pro 1,629
- 3 Zerank 1 1,573
- 4 Voyage AI Rerank 2.5 1,544
- 5 Zerank 1 Small 1,539
- 6 Voyage AI Rerank 2.5 Lite 1,520
- 7 Cohere Rerank 4 Fast 1,510
- 8
Qwen3-Reranker-8B 1,473
- 9 Contextual AI Rerank v2 Instruct 1,469
- 10 Cohere Rerank 3.5 1,451
- 11 BAAI/BGE Reranker v2 M3 1,327
- 12 Jina Reranker v2 Base Multilingual 1,327
Source: Official Agentset Reranker Leaderboard leaderboard
Bars start at 1,399.35.
- 1
Qwen3-Reranker-8B 1,473
Source: Official Agentset Reranker Leaderboard leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Zerank 2 | 1,638 | Official leaderboard | ||
| 2 | Cohere Rerank 4 Pro | 1,629 | Official leaderboard | ||
| 3 | Zerank 1 | 1,573 | Official leaderboard | ||
| 4 | Voyage AI Rerank 2.5 | 1,544 | Official leaderboard | ||
| 5 | Zerank 1 Small | 1,539 | Official leaderboard | ||
| 6 | Voyage AI Rerank 2.5 Lite | 1,520 | Official leaderboard | ||
| 7 | Cohere Rerank 4 Fast | 1,510 | Official leaderboard | ||
| 8 | | 1,473 | May 2025 | Official leaderboard | |
| 9 | Contextual AI Rerank v2 Instruct | 1,469 | Official leaderboard | ||
| 10 | Cohere Rerank 3.5 | 1,451 | Official leaderboard | ||
| 11 | BAAI/BGE Reranker v2 M3 | 1,327 | Official leaderboard | ||
| 12 | Jina Reranker v2 Base Multilingual | 1,327 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 1,473 | May 2025 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Agentset Reranker Leaderboard leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Knowledge
- MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects.
Intern-S2-397B 89.8%
- MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy.
Seed1.8 92.3%
- SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy.
Qwen 3.6 Max (Preview) 73.9%
- WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1.
Tencent Hy4 preview 83.9%
- BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly.
Gemma 4 31B IT 74.4%
- HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 18 months ago, in May 2025, for Qwen3-Reranker-8B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (12 entries).
- Pass Models scored 1 model we track, plus 11 other entries on its leaderboard.
- Warning Newest score The newest came 18 months ago, in May 2025, for Qwen3-Reranker-8B.
- Note Frontier models None of its models are on the capabilities index.
- Note Headroom Scored as an Elo rating, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 1 row for models we track and 11 other entries at the last read.
- Pass Last read November 18th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.