Use vLLM Reranker to reorder documents after retrieval from a vector store by relevance to the given query.
npm install n8n-nodes-reranker-vllmUse vLLM Reranker to reorder retrieved documents by their relevance to a given query.
This node leverages locally served reranker models (via vLLM to perform _cross-encoder–style_ relevance scoring.
> Built on the official n8n Community Node Starter, so you can develop, lint, and ship confidently.
---
This node currently supports Qwen3 Rerankers by @Qwen:
- Qwen/Qwen3-Reranker-8B
- Qwen/Qwen3-Reranker-4B
- Qwen/Qwen3-Reranker-0.6B
All variants and quantizations (e.g., :Q5_K_M, :Q8_0) are automatically detected.
If you’d like to request support for other reranker models, please open a GitHub issue or feature request so it can be added to the whitelist.
---
You can install this node directly from the n8n Community Nodes interface:
1. Go to Settings → Community Nodes in your n8n instance.
2. Enable _Community Nodes_ if you haven’t already.
3. Enter the package name:
4. Confirm and install.
---
1. Add the Reranker vLLM node to your workflow.
2. Connect it after a retriever or vector search node (e.g., Qdrant, Pinecone, Weaviate, etc.).
3. Provide:
- Query text – the user query or search question.
- Documents array – the list of retrieved text chunks.
4. Choose a supported Qwen3 Reranker model.
5. The node outputs documents reordered by their relevance scores (0–1 scale).
---
- https://medium.com/@kimdoil1211/deploying-qwen3-reranker-8b-with-vllm-instruction-aware-reranking-for-next-generation-retrieval-c35a57c9f0a6
Contributions and new model requests are welcome!
If you’d like to see support for another reranker model, please:
- Open a GitHub issue, or
- Submit a pull request with your proposed model configuration.