Canonical guide · Source docs/vps-and-n8n.md

VPS and n8n

Relmio can add either OAuth companion beside an existing n8n VPS. Each companion has its own provider sign-in, Docker volume, endpoint, and n8n credential. SuperGrok does not require or read ChatGPT credentials.

Connection	Private Base URL	n8n API-key field	Use Responses API
OpenAI OAuth with ChatGPT/Codex sign-in	`http://n8n-openai-oauth:10531/v1`	`local-only` placeholder	On in the Relmio OpenAI Chat Model v1.3 recipe
SuperGrok OAuth	`http://n8n-supergrok:14502/v1`	One-time local Relmio bearer	Off for workflow model nodes and Chat Hub

The OpenAI bridge is unofficial, private, and policy-uncertain. ChatGPT sign-in is not an OpenAI Platform API key. The SuperGrok adapter is experimental and uses a fresh official Grok device sign-in.

Relmio writes only below /docker/n8n-openai-oauth. It does not edit, rebuild, recreate, stop, or restart your n8n Compose project or image. Neither companion publishes a host port.

Use the wizard

Start Relmio and choose the provider. Complete ChatGPT/Codex sign-in for the OpenAI bridge, or choose SuperGrok and complete its device sign-in after the companion is installed.
Compare the shown SSH host fingerprint with your provider before you enter a password.
Select a running n8n container and one existing Docker network.
Review the plan and confirm it before Relmio writes anything remotely.
In n8n, use the matching private Base URL and Responses API setting from the table above, never 127.0.0.1.
Select Disconnect from VPS when you finish the remote setup.

Relmio also closes the authenticated SSH session after 15 minutes of inactivity. An active VPS operation holds a bounded lease so discovery or an approved install can finish before the idle timer resumes.

Update an existing OpenAI bridge

Installing a newer Relmio package on your computer does not replace the bridge already running on the VPS. Start a Relmio release that contains the bridge compatibility update, then use the same browser wizard:

Run relmio vps and reconnect to the VPS.
Compare and confirm its SSH host fingerprint.
Select the n8n container and Docker network.
Choose OpenAI-OAuth/Codex bridge, then select Manage OpenAI-OAuth/Codex bridge.
Choose Review bridge update and review the exact sidecar-only plan.
Select the confirmation checkbox, then choose Update the bridge.

Relmio performs the SSH update from the browser flow, so no separate VPS terminal is required. It uploads the current local ChatGPT sign-in, keeps the selected network, rebuilds and verifies only the owned sidecar inside /docker/n8n-openai-oauth, and publishes no host port. n8n remains untouched.

Use local-only only for the OpenAI bridge. It is a placeholder, not an OpenAI Platform API key. SuperGrok uses the one-time local bearer shown by its wizard.

Next guides

GPT Image 2.5 in n8n

A new OpenAI OAuth bridge includes gpt-image-2.5-flare and gpt-image-2.5-sunburst in model discovery. For an existing managed bridge, use its reviewed browser runtime-update action to install the current adapter; upgrading only the Relmio dashboard does not update an already running bridge. The local update preserves its saved sign-in. A VPS update follows its separate reviewed sign-in upload flow. Both paths leave n8n unchanged.

The completion screen shows image IDs from the verified model response. In n8n's OpenAI node, select Image, then Generate an Image or Edit Image, and pick the exact model from the list. Use By ID when needed by your n8n version. Existing gpt-image-2 remains available. Do not use the generic gpt-image-2.5 as a model ID or select an image model for a text-chat recipe.

Both variants passed bounded generation and editing tests. Exact output size and all options remain unverified; begin with low quality. Discovery is not a promise of account entitlement. Live, Realtime and audio are separate capabilities and remain unsupported through this bridge.