What It Is

RAG Fortress is a plug-and-play Retrieval-Augmented Generation system with support for multiple LLMs, vector stores, embedding providers, and custom security layers. It features semantic caching for up to 80% cost reduction, message encryption at rest, and unified configuration for effortless provider switching.

Open source, local-LLM friendly, and production-ready with enterprise security. Build the RAG system you need, not the one a vendor wants you to use.

Why It's Not Just Another RAG Tool

🔌

Vendor Agnostic by Default

Choose your LLM, your embeddings, your vector store, and your database. Nothing is locked in.

🏠

Internal LLM Support

Local models via llama.cpp or any self-hosted provider. No document leaves your environment if you don't want it to.

🔒

Enterprise-Grade Security

HTTPOnly cookie authentication, message encryption at rest, automatic log sanitization, and multi-level security clearance enforcement.

⚡

Intelligent Semantic Caching

Multi-tier semantic cache reduces LLM API costs by up to 80% and delivers instant responses for similar queries.

🎯

Unified Configuration

Switch between LLMs, embeddings, and vector stores with unified settings. Supports hybrid search, fallback LLMs, and system diagnostics for complete control.

👥

Designed for Real Teams

User management, admin dashboard, optional error reporting, logging, and organization support.

Core Features

✓ Easy setup with environment-based configuration

✓ Document ingestion with multiple file formats

✓ Supports OpenAI, Gemini, Llama.cpp, and HuggingFace models

✓ Plug-and-play vector store integrations (Qdrant, Pinecone, FAISS, Chroma)

✓ Multi-tenant ready with department isolation

✓ Role-based access control (RBAC)

✓ Chat interface + search interface

✓ Admin controls for LLM config, embeddings, vector store

✓ Semantic caching with RedisVL for 80% cost reduction

✓ Message encryption at rest with optional cache encryption

✓ HTTPOnly cookie authentication & automatic log sanitization

✓ Hybrid search (vector + keyword) with configurable weights

✓ System diagnostics & health monitoring

✓ Unified configuration for effortless provider switching

✓ RESTful API endpoints for integration

✓ Document approval workflow

✓ Demo mode for public showcases

Who Benefits from RAG Fortress?

From developers to enterprises, RAG Fortress adapts to your specific knowledge management needs.

👨‍💻

Developers & Engineers

• Spin up RAG systems without vendor lock-in
• Experiment with different LLM architectures
• Build custom RAG solutions for clients

🏢

Enterprises & Businesses

• Secure internal knowledge base with RBAC
• Department-level access control
• Policy & compliance document management

🎓

Educational Institutions

• Course materials & lecture notes hub
• Student AI study assistants
• Research paper repository with semantic search

🔐

Security-Conscious Organizations

• 100% on-premise deployment with local LLMs
• Air-gapped environment support
• Multi-level security clearance enforcement

🌍

NGOs & Nonprofits

• Community knowledge systems
• Resource libraries without cloud costs
• Multi-language document support

⚖️

Legal & Compliance Teams

• Case law & regulation research
• Contract & agreement analysis
• Audit trail & activity logging

How It Works

1

Upload Documents

Add PDFs, text files, CSVs, and more to your knowledge base

→

2

Configure Providers

Choose your LLM, embeddings, and vector store

→

3

Index Content

Process and vectorize your documents

→

4

Start Querying

Ask questions and get context-aware answers

See It In Action

Experience the power of intelligent document conversations

💬 RAG Fortress Chat

Intelligent Conversations with Your Documents

Get context-aware answers powered by your knowledge base

Try Live Demo

💬

Natural Conversations

Chat naturally with your documents and get intelligent responses

🔐

Enterprise Security

HTTPOnly cookies, E2E encryption, log sanitization & RBAC

⚡

Lightning Fast

Semantic caching reduces costs by 80% with instant responses

🎯

Context-Aware

Understands your documents and provides relevant answers

🤖

Multiple LLMs

Support for OpenAI, Gemini, HuggingFace, and local models

📊

Activity Tracking

Complete audit trails and analytics dashboard

Complete Platform Overview

Admin Dashboard

Activity monitoring & analytics

Application Settings

Configure LLMs & vector stores

Department Management

Organize teams & access control

Multi-Tier Invitations

Email-based user onboarding

Activity Logger

Detailed audit logs

Detailed Configuration

Fine-tune every setting

Quick Start Guide

Get RAG Fortress running in minutes with our streamlined setup process.

🔧 Backend Setup

# Clone repository
git clone https://github.com/nurudeen19/rag-fortress.git
cd rag-fortress/backend

# Install with uv (recommended)
uv sync

# Activate environment
.venv\Scripts\Activate  # Windows
# source .venv/bin/activate  # macOS/Linux

# Configure environment
cp .env.example .env
# Edit .env with your API keys

# Initialize database
python setup.py

# Start server
python run.py

🎨 Frontend Setup

# Navigate to frontend
cd frontend

# Install dependencies
npm install

# Configure environment
cp .env.example .env
# Set API_URL if needed

# Start development server
npm run dev

# Access at:
# Frontend: http://localhost:5173
# Backend: http://localhost:8000
# API Docs: http://localhost:8000/docs

🔑 Default Credentials

username:

admin

Password:

admin@RAGFortress123

📚 Documentation

Ready to Build Your Knowledge Platform?

Start building with RAG Fortress today. Open source, flexible, and built for teams.

Get Started on GitHub Try the Demo

A flexible, secure,vendor-agnosticknowledge platform.