Qwen3-Reranker-0.6B
Tiny 0.6B open cross-encoder reranker with 32K context
About
Qwen3-Reranker-0.6B is the smallest member of Alibaba's Qwen3 reranking family: a 0.6B-parameter cross-encoder that scores query-document pairs across 100+ languages with a 32K-token context window — unusually long for a reranker this small. It is instruction-aware, Apache 2.0 licensed, and Alibaba reports strong results on MTEB-R, CMTEB-R, and MLDR retrieval benchmarks. Released alongside the Qwen3 embedding series on June 5, 2025, it is meant to pair with Qwen3-Embedding models in a fully open retrieve-then-rerank stack. The tradeoff is the obvious one: a 0.6B cross-encoder will not match the ranking quality of 7B rerankers or Cohere's commercial models on hard queries — it buys you speed and cost, not the top of the leaderboard.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Strong MTEB-R / CMTEB-R / MLDR results (Alibaba-reported)
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (verified tier, metrics-v1). Evidence: 0.6B params, 32K context, 100+ languages, Apache 2.0, instruction-aware cross-encoder verified in Qwen's official model table and SiliconFlow's model page read this session.; June 5, 2025 release date from the Qwen3 Embedding series launch.; Quality-vs-size tradeoff vs larger rerankers stated honestly..