Langfuse integration

Questions
1

I want to integrate langfuse to my workflow for monitoring and observability. But apparently n8n dont provide native integrations to langfuse. Can anyone help me with this or provide a link to a resource that will help me integrate langfuse to n8n

1 Like
2

Hello @Derrick_Mureithi

You’re right, from what I see n8n doesn’t currently offer a native integration for Langfuse. However, you can still integrate Langfuse into your workflows by using its API via the HTTP Request node in n8n.

Here’s a general approach to help you get started:

  1. Set up your Langfuse API key (check your Langfuse dashboard).
  2. Use the HTTP Request node in n8n to send logs, traces, or events to Langfuse’s API endpoints.
  3. Format your requests to match the Langfuse API documentation (they support observability for LLM-based apps, traces, spans, evaluations, etc.).
  4. Optionally, you can build a simple custom function or use the Function node in n8n to format trace data before sending it.

Alternatively, if you’re comfortable with TypeScript and want a more streamlined integration, you could also consider building (coding) a custom n8n node for Langfuse. This would:

  • Let you reuse Langfuse actions easily across workflows
  • Provide a cleaner UI for configuration
  • Be contributed back to the community

Here’s a great starting point for building custom nodes: n8n Docs – Creating Your Own Node

1 Like
3

hello @Derrick_Mureithi,

I had the same issue recently, and found a way to connect Langfuse using the Langfuse code node.

I made this simple example workflow to show how it works:

Hope it helps!

1 Like
4

A community node would greatly benefit many users who are serious about tractability and improving based on feedback. Currently, it’s just not nice to iterate on LLM outputs with n8n nor to build workflows that need “production grade” quality as you can’t reasonably trace back defects/anomalies if they occur.