How to Set Up Odysseus AI

Updated June 2026 - Based on Odysseus v1.x

This guide gives you a quick overview of each setup method. For the most up-to-date instructions, always refer to the official GitHub repository .

Prerequisites

Recommended

The easiest way to get started. Docker handles all dependencies.

# Clone the repository

git clone https://github.com/pewdiepie-archdaemon/odysseus.git

cd odysseus

# Start with Docker Compose

docker compose up -d

# Open in browser

# http://localhost:8080

For GPU support, use docker compose -f docker-compose.gpu.yml up -d if available.

# Install dependencies

brew install python@3.11 git

# Clone and set up

git clone https://github.com/pewdiepie-archdaemon/odysseus.git

cd odysseus

python3.11 -m venv venv

source venv/bin/activate

pip install -r requirements.txt

# Start the server

python main.py

Apple Silicon Macs can use Metal acceleration with supported backends.

# Install Python 3.11+ and Git

# Then clone and set up

git clone https://github.com/pewdiepie-archdaemon/odysseus.git

cd odysseus

python -m venv venv

# Windows: venv\Scripts\activate

# Linux: source venv/bin/activate

pip install -r requirements.txt

python main.py

For NVIDIA GPU support, install CUDA drivers before running. Check the official docs for GPU-specific instructions.

Odysseus supports multiple model backends. The most common setup is Ollama for local models:

# Install Ollama (separate from Odysseus)

# See https://ollama.com/download for your platform

# Pull a model

ollama pull llama3.1:8b

# Odysseus will auto-detect Ollama running on localhost

Other options: vLLM, llama.cpp, or cloud APIs via OpenRouter/OpenAI.

For the latest installation instructions, always check the official README on GitHub. The project is actively evolving.