How to Intermediate · 4 min read

How to use RetrieverQueryEngine in LlamaIndex

Q: How to use RetrieverQueryEngine in LlamaIndex

Use RetrieverQueryEngine in llamaindex to perform efficient retrieval-augmented querying by combining a retriever with a query engine. Initialize a retriever from your index, then create a RetrieverQueryEngine with it to run queries that fetch relevant documents before answering.

Quick answer

Use RetrieverQueryEngine in llamaindex to perform efficient retrieval-augmented querying by combining a retriever with a query engine. Initialize a retriever from your index, then create a RetrieverQueryEngine with it to run queries that fetch relevant documents before answering.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install llamaindex openai

Setup

Install llamaindex and set your OpenAI API key as an environment variable.

bash

pip install llamaindex openai

Step by step

This example shows how to create a RetrieverQueryEngine from a GPTVectorStoreIndex and query it.

python

import os
from llamaindex import GPTVectorStoreIndex, SimpleDirectoryReader, RetrieverQueryEngine

# Load documents from a local directory
documents = SimpleDirectoryReader('data').load_data()

# Create an index from documents
index = GPTVectorStoreIndex.from_documents(documents)

# Create a retriever from the index
retriever = index.as_retriever()

# Initialize RetrieverQueryEngine with the retriever
query_engine = RetrieverQueryEngine(retriever=retriever)

# Query the engine
response = query_engine.query('What is the main topic of the documents?')
print(response.response)

output

The main topic of the documents is ...

Common variations

Use different index types like GPTListIndex or GPTTreeIndex with retrievers.
Customize retriever parameters such as search_kwargs for fine-tuning retrieval.
Use async querying by calling await query_engine.aquery(...) in async contexts.
Switch to other LLMs by passing a custom llm parameter to the index or query engine.

Troubleshooting

If queries return irrelevant results, increase the retriever's k parameter to fetch more documents.
Ensure your OpenAI API key is set correctly in os.environ['OPENAI_API_KEY'].
Check that your documents are loaded properly and the index is built without errors.
For slow queries, consider caching the index or using a smaller model.

✅

Key Takeaways

Use RetrieverQueryEngine to combine retrieval with LLM querying for efficient document QA.
Create a retriever from your index and pass it to RetrieverQueryEngine for best results.
Adjust retriever parameters like number of documents retrieved to improve answer relevance.

Verified 2026-04 · gpt-4o

Verify ↗