Get Started

Requirements

Python 3.10 or 3.11

To get started with the GraphRAG system, you have a few options:

👉 Install from pypi.
👉 Use it from source

Top-Level Packages

Indexing Pipeline Overview
Query Engine Overview

Overview

The following is a simple end-to-end example for using the GraphRAG system. It shows how to use the system to index some text, and then use the indexed data to answer questions about the documents.

Install GraphRAG

pip install graphrag

Running the Indexer

Now we need to set up a data project and some initial configuration. Let's set that up. We're using the default configuration mode, which you can customize as needed using environment variables or using a config file.

First let's get a sample dataset ready:

mkdir -p ./ragtest/input

Now let's get a copy of A Christmas Carol by Charles Dickens from a trusted source

curl https://www.gutenberg.org/cache/epub/24022/pg24022.txt > ./ragtest/input/book.txt

Next we'll inject some required config variables:

Set Up Environment Variables

First let's make sure to setup the required environment variables:

• GRAPHRAG_API_KEY - API Key for executing the model, will fallback to OPENAI_API_KEY if one is not provided.
• GRAPHRAG_LLM_MODEL - Model to use for Chat Completions.
• GRAPHRAG_LLM_MODEL_SUPPORTS_JSON - This will signal to the indexing engine that you're using a model capable of JSON-mode output (e.g. gpt-4 or gpt-3.5-turbo). We highly recommend enabling this to avoid malformed JSON errors during indexing.
• GRAPHRAG_EMBEDDING_MODEL - Model to use for Embeddings.
• GRAPHRAG_INPUT_TYPE - Type of input data, can be text or csv.
• GRAPHRAG_API_BASE - Base URL for the Azure OpenAI. Only required for Azure OpenAI users.
• GRAPHRAG_LLM_DEPLOYMENT_NAME - Deployment name for the Chat Completions model. Only required for Azure OpenAI users.
• GRAPHRAG_EMBEDDING_DEPLOYMENT_NAME - Deployment name for the Embeddings model. Only required for Azure OpenAI users.

OpenAI and Azure OpenAI

To get started, let's set the base environment variables.

export GRAPHRAG_API_KEY="<api_key>" && \
export GRAPHRAG_LLM_MODEL="<chat_completions_model>" && \
export GRAPHRAG_LLM_MODEL_SUPPORTS_JSON="True" && \
export GRAPHRAG_EMBEDDING_MODEL="<embeddings_model>" && \
export GRAPHRAG_INPUT_TYPE="text"

Azure OpenAI

In addition, Azure OpenAI users should set the following env-vars.

export GRAPHRAG_API_BASE="https://<domain>.openai.azure.com" && \
export GRAPHRAG_API_VERSION="2024-02-15-preview" && \
export GRAPHRAG_LLM_API_TYPE = "azure_openai_chat" && \
export GRAPHRAG_LLM_DEPLOYMENT_NAME="<chat_completions_deployment_name>" && \
export GRAPHRAG_EMBEDDING_API_TYPE = "azure_openai_embedding" && \
export GRAPHRAG_EMBEDDING_DEPLOYMENT_NAME="<embeddings_deployment_name>"

For more details about configuring GraphRAG, see the configuration documentation. For more details about using the CLI, refer to the CLI documentation.

Running the Indexing pipeline

Finally we'll run the pipeline!

python -m graphrag.index --root ./ragtest

This process will take some time to run. This depends on the size of your input data, what model you're using, and the text chunk size being used (these can be configured in your .env file). Once the pipeline is complete, you should see a new folder called ./ragtest/output/<timestamp>/artifacts with a series of parquet files.

Using the Query Engine

Running the Query Engine

Now let's ask some questions using this dataset.

Here is an example using Global search to ask a high-level question:

python -m graphrag.query \
--data ./ragtest/output/<timestamp>/artifacts \
--method global \
"What are the top themes in this story?"

Here is an example using Local search to ask a more specific question about a particular character:

python -m graphrag.query \
--data ./ragtest/output/<timestamp>/artifacts \
--method local \
"Who is Scrooge, and what are his main relationships?"

Please refer to Query Engine docs for detailed information about how to leverage our Local and Global search mechanisms for extracting meaningful insights from data after the Indexer has wrapped up execution.