DocuMind (RAG based desktop app)

👉 DocuMind: Your documents, your AI-powered mind. 🌿

Turn your documents into dynamic knowledge sources with DocuMind! Simply upload a document (currently supports only pdf format), ask questions, and get instant, accurate responses. Powered by advanced Retrieval-Augmented Generation (RAG) technology, DocuMind understands the content and provides clear and insightful answers. Whether it’s contracts, research papers, reports, or technical manuals — DocuMind helps you access information in seconds.

Features

Ask questions about your documents and get instant, accurate responses.
Provides relevant, fact-based answers using RAG.
Documents are processed securely on your local machine with no data leakage.

TechStack

Using Axum rust server at backend to expose REST Apis and connect with Ollama server for inference & Qdrant vector database for storage.
Tauri app with React UI for frontend.

🛠️Installation Setup

Pre-requisites

Make sure you have following installed

Rust
Ollama
Docker
Node (v18.19 or later) & yarn

Setup the Qdrant vector database

docker volume create qdrant_data

docker run -d \ --name qdrant \ -p 6333:6333 \ -v qdrant_data:/qdrant/storage \ qdrant/qdrant

Pull the AI models on your machine via Ollama CLI

ollama pull nomic-embed-text:latest
ollama pull llama3.1:8b

Clone the repository and build the application

# Clone the repo
git clone https://github.com/Harry-027/DocuMind
Change the directory
cd DocuMind
Run the server
make app_server
Run the tauri client in a new terminal
make tauri_client
Run the app client in another new terminal
make app_client

⚙️ Configuration

In the env.yaml, you can configure the following -

Chunking Strategy: You can configure the chunk size (embeddingmodelchunk_size).
Embedding Model: Customize the embedding model (embeddingmodelname) for better document understanding.
LLM Model: Customize the LLM model (generatemodelname) for better document understanding.