Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/pewdiepie-archdaemon/odysseus/llms.txt

Use this file to discover all available pages before exploring further.

Odysseus is a self-hosted AI workspace that runs entirely on your own hardware. It gives you a ChatGPT-style interface over local models (Ollama, vLLM, llama.cpp) and cloud APIs (OpenAI, OpenRouter, GitHub Copilot) — with no data sent anywhere you don’t control. Beyond chat, Odysseus includes an autonomous agent, a one-click model manager (Cookbook), deep research, a document editor, email, calendar, persistent memory, and more.

Quick Start

Get Odysseus running in minutes with Docker or native Python.

Configuration

Explore every environment variable and in-app setting.

Features

Chat, Agent, Cookbook, Deep Research, Documents, and more.

Deployment

Docker, Linux, macOS, Windows, GPU setup, and HTTPS.

Integrations

Connect local models, cloud providers, MCP servers, and email.

Security & Admin

Users, auth, API tokens, and hardening for network deployments.

​
What’s Inside

Odysseus ships as a single FastAPI application with a modular JavaScript frontend. Every feature is self-contained and configurable from the Settings panel — no YAML sprawl required.

Chat

Streaming chat with any OpenAI-compatible endpoint. Presets, file uploads, vision, and session history.

Agent

Autonomous multi-round agent loop with web, shell, file, and MCP tools.

Cookbook

Scan your GPU, browse compatible models, download and serve with one click.

Deep Research

Multi-step iterative research that searches, reads, and synthesizes sources into a visual report.

Documents

Multi-tab editor for Markdown, HTML, and CSV with inline AI assistance.

Email & Calendar

IMAP/SMTP inbox with AI triage and CalDAV calendar sync.

Memory & Skills

Persistent vector memory and reusable agent skills powered by ChromaDB.

Notes & Tasks

Quick notes with reminders, a todo list, and cron-style scheduled tasks.

​
Get Started

1

Clone and start with Docker

git clone https://github.com/pewdiepie-archdaemon/odysseus.git
cd odysseus
cp .env.example .env
docker compose up -d --build
2

Open the app

Navigate to http://localhost:7000. On first boot, Odysseus prints a temporary admin password in the Docker logs — grab it with docker compose logs odysseus | grep -i password.
3

Add a model provider

Go to Settings → Models and add your first provider. For a fully local setup, point Odysseus at Ollama (http://localhost:11434/v1). For cloud access, paste an OpenAI or OpenRouter API key.
4

Start chatting

Open the Chat tab, pick a model from the dropdown, and send your first message. From there, explore Agent mode, Cookbook, and the rest of the workspace.
Odysseus defaults to binding on 127.0.0.1 — it is only accessible from your local machine. See HTTPS & Network Exposure before opening it to a LAN or the internet.

Build docs developers (and LLMs) love

Get started for freeTalk to us