Enigma LLM

This LLM-RAG project uses Next.js + shadcn-ui + python(flask + chromadb). Aim to create a lightweight RAG framework.

Key Features

• Document Upload & Management: Easily upload personal documents (markdown format) and manage/delete them within the interface.
• Vectorization & Storage: Utilizes a Python backend (Flask, ChromaDB) to convert documents into vectors and store them in a vector database for efficient semantic search.
• Question Answering: Ask questions in natural language, and the system will generate relevant and informative answers based on your documents, powered by an LLM (Large Language Model).
• User-Friendly Interface: Built with Next.js and Shadcn UI for an intuitive and aesthetically pleasing experience.
• Customization: Adjust LLM model, vectorization methods, and other settings to suit your needs.

Getting Started

Note

You need to run the llama.cpp server You can refer to here for more information We consider to add a local start service in future.

First, run the development server:

npm install
npm run dev

Second, run the data process server:

cd data_process_module
conda create -n llm_ta python==3.11
conda activate llm_ta
pip install -r requirements.txt

Finally, run the backend server:

python main.py

Open http://localhost:3000 with your browser to see the result.

Usage

You can add your personal documents in documents page, and ask question in Home page