Skip to main content
n8n is a workflow automation platform with AI nodes built in. Opper ships a verified community node, so Opper models appear as a native chat-model sub-node rather than a generic HTTP request. Requirements: n8n 1.94.0 or later, Node.js 20.15+.

​
Setup

1

Install the node

Open the nodes panel on the canvas (+, or press n) and search for Opper. Verified community nodes appear under More from the community at the bottom of the panel. Select Opper Chat Model, then select Install.Only instance owners and admins can install; once installed, every member can use it in their workflows.The npm package behind it is @opperai/n8n-nodes-opper, if you would rather install it from Settings → Community Nodes.
2

Add the credential

Create an Opper API credential:The credential test validates connectivity through Opper’s OpenAI-compatible GET /models endpoint.
3

Use it in a workflow

Add an AI Agent node (or Basic LLM Chain, or anything else that accepts a language-model connection) and attach Opper Chat Model as its model sub-node.An n8n canvas with Opper Chat Model connected to the Chat Model input of an AI Agent node, and the nodes panel listing Opper Chat Model alongside the built-in providers

​
Choosing a model

The Opper Chat Model node lists the catalog fetched with your credential, filtered by your model access rules. Agent nodes need tool calling — pick a model the catalog marks with the tools capability. The Opper Chat Model parameters panel with an Opper credential selected and the Model dropdown open on EU-hosted evroc models Switching models is a dropdown change rather than a rebuild, so the same workflow can run on Claude, GPT, Gemini, Mistral or an EU-sovereign open-weight model. Picking a model such as evroc/zai-org/GLM-5.2 keeps inference inside the EU without touching anything else in the workflow. Under Options you can set temperature, top-p, frequency and presence penalties, max tokens, timeout, retries, and a JSON response format. Every call is traced in the Control Plane with its token usage and cost.

​
Attributing usage per workflow

n8n instances usually run many workflows against one credential. Tag the traffic so you can tell them apart in Opper analytics — see Tags and usage attribution.

​
Troubleshooting

n8n loads community nodes at startup. Restart the instance after installing.
Verified community nodes can be hidden per instance. On n8n Cloud, an instance owner toggles them in the Cloud Admin Panel; self-hosted instances control it with environment variables. Note also that only owners and admins can install a node — members can only use one that is already installed.
It calls GET /models. A failure usually means the key is wrong, or the base URL has an extra path segment — it should be exactly https://api.opper.ai/v3/compat.

Tags and usage attribution

Split usage per workflow or tenant.

All integrations

Every app that runs on Opper.