Model•active
Qwen3-Reranker-0.6B
Qwen3-Reranker-0.6B is a 0.6B-parameter open-weight cross-encoder reranking model for multilingual and code retrieval.
Official siteVerified Oct 4, 2026
Overview
The model accepts query-document text pairs and returns a relevance score. It is the compact representative of Qwen3 Reranker and supports a 32K sequence length.
Model specifications
- Family
- Qwen3 Reranker
- Context
- 32,000 tokens
- Input
- Text
- Output
- Score
- Open weights
- Yes
- Capabilities
- —