TALLA-RAG

Table of Contents

1. About The Project
   • Built With
2. Usage
3. Getting Started
   • Prerequisites
   • Installation
4. Roadmap
5. Contributing
6. License
7. Contact
8. Acknowledgments

About The Project

Built With

(back to top)

Usage

TALLA-RAG is a local application that lets you chat with your documents using a local LLM backed by a Cassandra cluster for vector storage.

1. Upload a .txt or .pdf file via the sidebar.
2. Click Ingest to Cluster — chunks are embedded and stored across Cassandra nodes.
3. Ask questions in the chat — the app retrieves relevant chunks and answers using the LLM only.
4. Use Clear All Knowledge to wipe the vector store.

Getting Started

Prerequisites

• Docker & Docker Compose
• Ollama running locally with the following models pulled:

  ollama pull nomic-embed-text:v1.5
  ollama pull granite3.2:2b
  ollama pull <your_model_name>

• Python 3.11 or higher
• uv (Python package manager)

     pip install uv

Installation

1. Clone the repo

   git clone https://github.com/nnay29/cassandra-cluster-RAG.git
   cd cassandra-cluster-RAG

2. Copy and configure the environment file

   cp .env.example .env

   Edit .env and set DOCKER_HOST_IP to you machine's local IP adress

3. Start the Cassandra cluster

   docker-compose up -d

4. Create a virtual environment

   uv venv

5. Install Python dependencies

   uv sync

6. Run the Streamlit app

   streamlit run app.py

   The app will be available at http://localhost:8501

(back to top)

Roadmap

See the open issues for a full list of proposed features (and known issues).

(back to top)

Contact

Project Link: https://github.com/nnay29/cassandra-cluster-RAG

(back to top)

Acknowledgments

• Working on uv projects
• Taximan Talla officiel

(back to top)