Local RAG Assistant with DeepSeek-R1

A locally-running Retrieval Augmented Generation (RAG) system that enables document analysis and question answering using DeepSeek.

Features

• Document Upload (PDF, DOCX, TXT)
• Interactive File Previews
• Local AI Processing with DeepSeek-r1
• Conversational Interface
• Vector Database Integration
• Cross-document Search Capabilities

Architecture

Usage

To install and use locally, the following steps could be followed

Prerequisites

1. Ollama Installation: Ollama install
2. Deepseek-r1 Installation: Deepseek download

Installation

Clone repository

git clone https://github.com/busayojee/RAG.git

Create virtual environment

python -m venv venv
source venv/bin/activate 

Install dependencies

pip install -r requirements.txt

• Update ModelFile to include deepseek-r1 path

Add to Ollama

ollama create deepseek-1.5 -f Modelfile

Start the application

streamlit run main.py

Contributing

Contributions are welcome! If you'd like to improve the app.

---