Phoenix has first-class support for LangChain applications.

To begin, you will have to start a Phoenix server. You can do this by running:

import phoenix as px
session = px.launch_app()

Once you have started a Phoenix server, You will instrument LangChain so that spans are created whenever you run a chain.

from phoenix.trace.langchain import LangChainInstrumentor

# By default, the traces will be exported to the locally running Phoenix 
# server. If a different endpoint is desired, change the environment
# variable os.environ["PHOENIX_COLLECTOR_ENDPOINT"] = ...

# Initialize your LangChain application

# Note that we do not have to pass in the tracer as a callback here
# since the above instrumented LangChain in it's entirety.
response =

By instrumenting LangChain, spans will be created whenevera a chain is run and will be sent to the Phoenix server for collection.

To view the traces in Phoenix, simply open the UI in your browser.


For a fully working example of tracing with LangChain, checkout our colab notebook.

Last updated