Model results
Reranker Simple Benchmark
Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.
Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.
Reranker Simple Benchmark
- Source
- Reranker Simple Benchmark (reviewed publisher results)
- Version
- 331162ca3ba8e4a139ee04195e48618e2976b198
- Measurement date
- Not reported
- Metric
- NDCG@10 · Higher is better
- Evaluation scope
- Official kMTEB nine-task Korean mean; gold-injected reranking with Mecab BM25 top-50 negatives; max_length 8192; standard task splits with MLDR dev+test average; excludes first-stage retrieval failures; publisher-rounded mean; result date not supplied.
Fewer than two comparable models. No ranking is assigned. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.
Qwen3 Reranker 8B
qwen/qwen3-reranker-8b
0.899NDCG@10
View modelEvidence independent