Integrations

LobeChat

Trace LobeChat conversations in Muster — set four env vars and toggle analytics in Settings.

LobeChat is an open-source ChatGPT-style UI. With Muster (langfuse) configured, every chat interaction is traced — usage, cost, and per-model performance.

Setup

1. Get Muster API keys

2. Set environment variables before launching LobeChat

ENABLE_LANGFUSE=1
LANGFUSE_SECRET_KEY=sk-lf-...
LANGFUSE_PUBLIC_KEY=pk-lf-...
LANGFUSE_HOST=https://app.getmuster.io

In Docker Desktop, add these as Environment variables when running the LobeChat container:

LobeChat env vars in Docker Desktop

3. Enable analytics in LobeChat

In LobeChat, go to Settings → About and enable analytics so traces are sent to Muster.

LobeChat settings — enable analytics

4. Use it

Configure your LLM credentials in LobeChat and start chatting.

LobeChat conversation

Traces appear in your Muster Traces table:

LobeChat trace example in Muster