LOLLMS RAG System

Overview

The LOLLMS RAG (Retrieval-Augmented Generation) System is a powerful tool that combines a FastAPI backend with a JavaScript client to provide efficient document management and search capabilities. This system allows users to add, remove, index, and search documents using vector embeddings, enhancing the retrieval process for various applications.

Components

Python Backend (lollms_rag.py): A FastAPI-based server that handles document management, indexing, and search operations.
JavaScript Client (LOLMSRAGClient): A client-side class that interacts with the backend API, providing an easy-to-use interface for developers.

Features

Secure authentication using bearer tokens
Document management (add, remove, index)
Vector-based document search
Database wiping capability
Cross-platform compatibility

Prerequisites

Python 3.7+
FastAPI
LOLLMS WebUI
Pydantic
lollmsvectordb
JavaScript environment (for client-side usage)

Installation

Just install, configure and run lollms and import your client inside your webapp. By default the url should be http://localhost:9600/lollms_js. If you have changed the host name or port address, just change this to whatever you have built. Once imported you are ready to use the rag functionalities in your lollms webapp.

Usage

Python Backend

Include the lollms_rag.py module in your FastAPI application.
Configure the LOLLMS WebUI instance with appropriate RAG settings.

Add the router to your FastAPI app:

from lollms_rag import router as rag_router

app.include_router(rag_router)

JavaScript Client

Include the LOLMSRAGClient class in your JavaScript project.

Create an instance of the client:

const ragClient = new LOLMSRAGClient('http://your-api-url', 'your-bearer-token');

Use the client methods to interact with the RAG system:

// Add a document
ragClient.addDocument('Title', 'Content', 'path/to/document')
  .then(response => console.log(response))
  .catch(error => console.error(error));

// Search for documents
ragClient.search('query text')
  .then(results => console.log(results))
  .catch(error => console.error(error));

API Endpoints

POST /add_document: Add a new document
POST /remove_document/{document_id}: Remove a document
POST /index_database: Index the database
POST /search: Search for documents
DELETE /wipe_database: Wipe the entire database

Security

The system uses bearer token authentication. Ensure that you keep your API keys secure and don't expose them in client-side code.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the Apache 2.0.

2.8 KiB Raw Permalink Blame History