How to beginner · 3 min read

How to use query engine in LlamaIndex

Q: How to use query engine in LlamaIndex

Use the llama_index library to build an index from your documents and then create a QueryEngine to run natural language queries against it. The query engine processes your input and returns relevant document snippets or answers.

Quick answer

Use the llama_index library to build an index from your documents and then create a QueryEngine to run natural language queries against it. The query engine processes your input and returns relevant document snippets or answers.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install llama-index openai

Setup

Install the llama-index package and set your OpenAI API key as an environment variable.

Install with pip install llama-index openai
Set environment variable in your shell: export OPENAI_API_KEY='your_api_key'

bash

pip install llama-index openai

Step by step

This example shows how to create a simple document index and query it using the QueryEngine in LlamaIndex.

python

import os
from llama_index import SimpleDirectoryReader, GPTVectorStoreIndex, QueryEngine

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

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

# Create a query engine from the index
query_engine = QueryEngine(index=index)

# Run a query
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

You can customize the query engine by using different index types like GPTListIndex or GPTTreeIndex, or by enabling streaming responses. Async usage is also supported.

python

from llama_index import GPTListIndex

# Build a list index instead
list_index = GPTListIndex.from_documents(documents)
query_engine = QueryEngine(index=list_index)
response = query_engine.query('Summarize the documents.')
print(response.response)

output

Summary of the documents: ...

Troubleshooting

If you get authentication errors, verify your OPENAI_API_KEY environment variable is set correctly.
If queries return empty or irrelevant results, ensure your documents are loaded properly and the index is built without errors.
Check your network connection if API calls fail.

✅

Key Takeaways

Use GPTVectorStoreIndex or other index types to build your document index.
Create a QueryEngine from the index to run natural language queries.
Set your OpenAI API key in os.environ before running queries.
Customize the query engine with different index types or async calls for flexibility.
Troubleshoot by verifying API keys, document loading, and network connectivity.

Verified 2026-04 · gpt-4o, claude-3-5-sonnet-20241022

Verify ↗