Page cover

🧩 Installation

Whether you're running it locally as a Python dev or deploying in containers as an ops engineer—Locentra OS installs cleanly, boots fast, and stays under your control.

This guide covers both manual setup and Docker-based deployment.


🧪 System Requirements

Minimum:

  • Python 3.10+

  • Node.js 18+

  • PostgreSQL 13+

  • Docker & Docker Compose (for containerized setup)

Recommended:

  • Unix-like system (Linux/macOS)

  • 8GB+ RAM for live inference and training

  • GPU (CUDA) if using large models locally


🧱 1. Clone the Repository

git clone https://github.com/Locentra/OS.git
cd OS

🐍 2. Manual Installation (Local Python Setup)

Ideal for devs who want to step through the system layer by layer.

🔹 Backend Setup

cd backend
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

🔹 Environment Variables

Copy the default .env and adjust settings:

cp .env.example .env

Example config:

MODEL_NAME=tiiuae/falcon-rw-1b
DATABASE_URL=postgresql://user:pass@localhost:5432/Locentra
DEBUG=true

🔹 Frontend Setup

cd ../web
npm install

🔹 Start Services

  • Backend:

    cd ../backend
    bash ../scripts/run_server.sh
  • Frontend:

    cd ../web
    npm run dev
  • Initialize Database:

    python ../database/db_init.py

Fastest way to run the full stack locally. Clean, reproducible, no system conflicts.

🔹 One-Liner Boot

cp .env.example .env
docker-compose up --build

🔹 Access Points

Logs are streamed to console by default. Extend docker-compose.yml to route logs or mount volumes as needed.


📁 4. Backend Structure Overview

Locentra-os/backend/
├── api/         # FastAPI routes (REST API layer)
├── core/        # Engine bootstrapping, registry, settings
├── models/      # Inference, model loader, fine-tuning, adapter
├── agents/      # AutoTrainer, FeedbackAgent (self-learning logic)
├── db/          # SQLAlchemy models, migrations, connectors
├── services/    # Logic layer: memory, analytics, user mgmt
├── data/        # Tokenizer, vector embedding, dataset prep
├── utils/       # Metrics, crypto helpers, evaluation

The backend is modular—every service, agent, or component can be extended or replaced.

Last updated