Multi-Agent LLM

Project Overview

This project implements a multi-document agent using the LlamaIndex framework. It is designed to manage document-specific agents that can perform QA and summarization tasks within individual documents, as well as a top-level agent that orchestrates queries across these document agents.

Features

• Document-Specific Query Handling: Each document has a dedicated agent capable of answering queries specific to its content.
• Top-Level Query Orchestration: A top agent manages interactions across multiple document agents.
• Reranking and Query Planning: Utilizes advanced features like reranking during document retrieval and dynamic query planning for efficient information retrieval.

Installation

Prerequisites

• Python 3.7 or higher
• pip

Setup Instructions

1. Clone the repository:

   git clone https://github.com/yourusername/llamaindex-multi-document-agent.git
   cd llamaindex-multi-document-agent

2. Install dependencies::

   pip install -r requirements.txt

Usage

To run the agent, navigate to the project directory and execute:

python main.py

Configuration

Modify the src/config.py to set the necessary API keys and parameters.