Installation

Vectra can be deployed quickly via Docker (recommended) or manually for development.

🚀 Quick Start (Docker)

The easiest way to launch Vectra in production or for a quick test.

1. Clone the project:

git clone https://github.com/HuguesGauthier/Vectra.git
cd Vectra

1. Configure environment:

cp .env.example .env
# Edit the .env file with your API keys (Gemini, etc.)

1. Launch services:

   docker-compose up -d
   

Access the UI at: http://localhost:9000

---

🛠️ Manual Installation (Development)

If you wish to modify the code or contribute to the project.

Prerequisites

• Python 3.10+
• Node.js (for the frontend)
• Databases: PostgreSQL, Qdrant, and Redis must be accessible.

Backend

1. Navigate to the backend folder:

   cd backend
   

2. Create a virtual environment:

   python -m venv .venv
   .venv\Scripts\activate  # Windows
   

3. Install dependencies:

   pip install -r requirements.txt
   

4. Launch the API:

   python main.py
   

Frontend

1. Navigate to the frontend folder:

   cd frontend
   

2. Install dependencies:

   npm install
   

3. Launch the development server:

   npm run dev