Qwen Qwen3 Reranker 4B

About this model

Qwen3 Reranker 4B is a 4B parameter text reranking model built on the Qwen3 dense foundation model, designed for high-quality passage reranking across 100+ languages. It supports a 32K context length and instruction-aware reranking for tailored retrieval scenarios. The model achieves 69.76 on MTEB-R, 75.94 on CMTEB-R, and 72.74 on MMTEB-R benchmarks. It is released under the Apache 2.0 license.

Technical specifications

Capabilities

Input modalities

Output modalities

Reasoning

No

Knowledge horizon