Nexus RAG is an enterprise-grade retrieval-augmented generation engine that runs entirely on your local machine. No data leaks, no subscriptions, just pure intelligence.
Modern AI engineering for local-first data processing.
Organize your knowledge into separate notebooks. Isolated, secure, and always relevant.
Powered by Ollama and HuggingFace, your data never leaves your hardware. Zero-cloud architecture.
Advanced document parsing and vector chunking using LangChain for optimal query context.
Powered by ChromaDB, your vector stores are persisted locally for instant availability.
Run NexusRAG on your own hardware in less than 5 minutes.
git clone https://github.com/artjomartur/nexus-rag.git
cd nexus-rag
python -m venv venv
source venv/bin/activate # macOS/Linuxpip install -r requirements.txt
cp .env.example .envpython main.pyOpen localhost:8000 to start querying.
Upload. Drag and drop your PDFs into a notebook.
Index. The engine automatically embeds your data locally.
Query. Ask anything and get context-aware answers instantly.
Interested in this project or looking for a developer? Send me a message.