Use Laneflip with n8n
n8n talks to OpenAI-compatible APIs through its OpenAI credential. Point that credential at Laneflip once and every OpenAI Chat Model and AI Agent node that uses it gets failover, hard spend caps and request logs, without changing the workflows.
Before you start
You need a Laneflip account with at least one lane and a Laneflip API key, and an n8n instance (self-hosted or cloud).
Set it up
- In n8n, create a new OpenAI credential.
- Paste your Laneflip API key into the API Key field. Never use a provider key here: your provider keys stay in Laneflip.
- Set Base URL to https://api.laneflip.com/v1 and save the credential. Unverified: n8n's documentation for the OpenAI credential lists only an API Key and an optional Organization ID and does not mention a Base URL field. Recent n8n versions show a Base URL field in the credential, but we have not yet confirmed this step with a real n8n instance. If your version has no such field, this setup does not apply to it.
- Add an OpenAI Chat Model node (or an AI Agent with an OpenAI Chat Model) and select this credential.
- Open the model list and pick one of your lanes. The list comes from Laneflip's models endpoint and shows your organisation's lanes. If it stays empty, switch the model field to enter the lane name by hand.
What you get centrally
- Failover: when a provider is slow, down or erroring, the request moves to the next approved model in the lane.
- Hard caps: requests stop at the spend cap you set on your own keys, in every workflow at once.
- Logs: every request from every workflow is logged in one place.
If something goes wrong
An invalid Laneflip key, a lane that does not exist or a reached cap comes back as an error message in the n8n node. Check the key and the lane name first, then the cap on your keys.