Deploy

Local deployment

Run Cradle Server on your own computer — one command, bound to localhost.

Everything on one machine: cradle-server runs on your computer and the console on opencradle.ai connects to it (topology T0). The step-by-step path, with screenshots, is Run Cradle on your machine; this page is the install command and the settings behind it.

[one machine]
  console (browser) ──API──► cradle-server (local GPU/CPU)

Install

You need Node 22, pnpm and curl (macOS or Linux). Rootless — nothing is written outside your home directory:

curl -fsSL https://download.opencradle.ai/cradle-cli/latest/cradle-cli-install.sh | CRADLE_BIND=127.0.0.1 sh
~/.local/bin/cradle-server --data-dir ~/.local/share/cradle/data --port 31416 --bind 127.0.0.1

The installer unpacks into ~/.local/share/cradle, bootstraps the database and saves the admin API key to ~/.local/share/cradle/data/client-credentials.json. The installer binds to 0.0.0.0 unless you set CRADLE_BIND=127.0.0.1 as above. Where a systemd user session exists, the installer also starts the server as a service, and the second line then reports that the port is already in use.

Local requests need the key too: paste it into the console when you press Connect on the This computer card.

Settings

Variable / flagDefaultMeaning
--port31416HTTP API port
--data-dir~/.cradledatabase, models, admin key
CRADLE_BIND / --bind127.0.0.1CRADLE_BIND=0.0.0.0 opens the server to your LAN

To expose it beyond the LAN, bind to 0.0.0.0 and put a reverse proxy (Caddy or nginx) in front for HTTPS — see remote-server deployment.

When to move off local

A local deployment is enough for a single operator and modest message volumes — the bottleneck is the model on the GPU, not the embedded SQLite database. Move to a dedicated GPU server when you need a shared box multiple clients connect to, or to a datacenter service when you need per-tenant isolation.