How to beginner · 4 min read

How to use SummaryIndex in LlamaIndex

Q: How to use SummaryIndex in LlamaIndex

Use SummaryIndex in LlamaIndex to build a concise summary representation of documents for efficient querying. Initialize the index with your documents, then query it with natural language prompts to get summarized answers.

Quick answer

Use SummaryIndex in LlamaIndex to build a concise summary representation of documents for efficient querying. Initialize the index with your documents, then query it with natural language prompts to get summarized answers.

PREREQUISITES

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

Setup

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

bash

pip install llama-index openai

Step by step

This example shows how to create a SummaryIndex from documents and query it for a summary answer.

python

import os
from llama_index import SimpleDirectoryReader, SummaryIndex, ServiceContext
from openai import OpenAI

# Set your OpenAI API key in environment variable before running
# export OPENAI_API_KEY="your_api_key"

# Initialize OpenAI client
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Load documents from a directory (replace 'data' with your folder)
docs = SimpleDirectoryReader('data').load_data()

# Create a service context with OpenAI client and model
service_context = ServiceContext.from_defaults(
    llm=client,
    chunk_size_limit=512
)

# Build the SummaryIndex
summary_index = SummaryIndex.from_documents(docs, service_context=service_context)

# Query the index
query = "Summarize the main points of the documents."
response = summary_index.query(query)

print("Summary answer:", response.response)

output

Summary answer: <concise summary text generated by the model>

Common variations

Use different LLM models by changing the llm parameter in ServiceContext.from_defaults().
Load documents from other sources like PDFs or URLs using appropriate loaders.
Use async calls if integrating in async frameworks by adapting the client calls accordingly.

Troubleshooting

If you get authentication errors, verify your OPENAI_API_KEY environment variable is set correctly.
If the summary is too short or irrelevant, increase chunk_size_limit or adjust your prompt.
For large documents, consider chunking or using other index types for better performance.

✅

Key Takeaways

Use SummaryIndex to create a compact summary representation of documents for efficient querying.
Initialize SummaryIndex with documents and a ServiceContext that wraps your LLM client.
Query the index with natural language prompts to get concise summaries.
Adjust chunk size and model parameters to optimize summary quality and performance.
Ensure your OpenAI API key is set in os.environ to avoid authentication issues.

Verified 2026-04 · gpt-4o

Verify ↗