Skip to content
OdyssAI

Getting started

PathYou needYou get
A — A CoeOS instanceA conversation with usYour dedicated CoeOS: the web app for your team, routing by skill, confidentiality by design
B — CoeOS DesktopA CoeOS instance, and a Mac, a Windows or a Linux machineThe agent on your machine, with full access, routed by your instance
C — The OdyssAI-X engineOne or more Apple Silicon MacsFrontier open-weight models served locally, OpenAI- and Anthropic-compatible

Path A — A CoeOS instance

Section titled “Path A — A CoeOS instance”

CoeOS is a managed service. We deploy and run a dedicated instance for you: the app, the CoeOS box, the database and the memory, on a machine of your own, shared with no other client.

  1. Pick a level. Open (all in the cloud, with or without DPA and ZDR), Hybrid (confidential work on a server on your premises) or Local (everything on site). See What CoeOS is.
  2. Write to us at hello@odyssai.eu. Pricing is on request.
  3. Your administrator receives the instance, creates the accounts, and chooses which providers are marked safe for confidential work.

Day-to-day use is documented on docs.coeos.io.

Path B — CoeOS Desktop

Section titled “Path B — CoeOS Desktop”

CoeOS Desktop connects to your instance’s box; it does not hold any provider key itself.

  1. Install CoeOS Desktop for your system (macOS, Windows or Linux). It updates itself afterwards.
  2. Connect it to your instance.
  3. Pick the permission mode: Ask (default, every command and write waits for you) or Grant all.

Full page: CoeOS Desktop.

Path C — The OdyssAI-X engine

Section titled “Path C — The OdyssAI-X engine”

One server (the orchestrator container; a Mac mini is enough) and one or more nodes (the Macs that hold the models; the server can be a node). The whole install is a terminal — or an AI agent following the repo’s AGENTS.md.

Terminal window
git clone https://github.com/Odyssai-eu/OdyssAI-X.git && cd OdyssAI-X
scripts/bootstrap-node.sh user@node.lan            # once per node, idempotent
scripts/wired-limit/install.sh user@node.lan 250880 # pin the GPU memory budget (needs the node's password)
mkdir -p ~/.odysseus && cp config/topology.example.yaml ~/.odysseus/topology.yaml
$EDITOR ~/.odysseus/topology.yaml                  # ssh targets, models_dir, backend
docker compose up -d
curl -s http://localhost:8000/health               # {"status":"idle","version":"…"}

Download a model from the dashboard (http://localhost:8000/, Models → Download), load it on a cluster, and chat on /v1/chat/completions. Single Mac? Use host.docker.internal as the SSH target; the example topology shows it. Full page: Install the engine.

Putting them together

Section titled “Putting them together”
CoeOS web app · CoeOS Desktop  ──▶  your CoeOS box  ──▶  cloud providers
                                                    └──▶  OdyssAI-X on your premises (Hybrid, Local)

For Hybrid and Local, the engine is linked to your box by an encrypted tunnel and declared in the box as a local provider, marked safe. Confidential conversations are then routed on the box’s Local table, built from safe models only. A replica pool is the natural backend: one copy of the model per Mac, many conversations per Mac at once.

What next

Section titled “What next”