# odysseus

> Self-hosted workspace for AI chat, agents, research, documents, email and calendar

- Repository: https://github.com/odysseus-dev/odysseus
- Category: Models & local AI
- Stars: 94,235 · Forks: 1,721 · Language: Python · License: AGPL-3.0
- Experience: Some experience
- Page: https://ir.adyog.com/repos/odysseus-dev/odysseus/

Odysseus is an AI workspace you run on your own machine or server. Instead of separate apps for chatting with a model, writing documents, reading email and keeping a calendar, you get them in one web interface with AI built into each part.

You start it with Docker Compose and open it in the browser at localhost:7000. It can use models you host locally or hosted APIs. Agents can use tools, MCP, files, a shell, skills and memory. A Cookbook section suggests models that fit your hardware, then downloads and serves them.

The README does not compare it to other projects by name. The difference is breadth: IMAP/SMTP email, CalDAV calendar sync, a writing-focused editor and blind model comparison all sit beside the chat. The README warns it has powerful local tools, so keep authentication on if it is reachable over a network.

## What it does

- **Chat and agents**: Works with local or API models. Agents can use tools, MCP, files, a shell, skills and memory.
- **Cookbook for local models**: Recommends models that suit your hardware, then handles downloading and serving them.
- **Deep Research**: Runs multi-step web research, reads the sources and produces a report.
- **Compare**: Tests models side by side in a blind format, then helps synthesize the answers.
- **Email, notes and calendar**: IMAP/SMTP inbox with triage, tags, summaries, reminders and reply drafts, plus todos, scheduled agent tasks and CalDAV sync.

## What you could build

- Run a private chat assistant on your own hardware with a local model.
- Ask an agent to research a topic on the web and write up a sourced report.
- Compare two or more models on your own prompts without knowing which is which.
- Triage an email inbox with AI summaries and reply drafts.
- Draft and edit documents with AI suggestions in a Markdown editor.

## Who it's for

Suits self-hosters and tinkerers who are comfortable with Docker and want one private AI workspace; know the basics of running containers and securing a web service first.

## Get started

Clone the repo, copy .env.example to .env, and start it with Docker Compose. Open http://localhost:7000 once the containers are healthy, and find the first admin password in the odysseus container logs.

```
git clone https://github.com/odysseus-dev/odysseus.git
```
```
cd odysseus
```
```
cp .env.example .env
```

Tags: self-hosted, ai-workspace, agents, local-models, docker