Qwen3-Reranker-4B
- Name in the API
Qwen/Qwen3-Reranker-4B- Base model
- Qwen/Qwen3-Reranker-4B
- Type
- Reranking
- Input → output
- a query and a list of documents → a relevance score per document
- Context length
- 40,960 tokens
- Availability
- Dedicated
- Hardware
- A 24 GB slice of a GPU per instance
- Endpoint
/v1/rerank