LangSmith vs Langfuse: which LLM observability platform should you use?

import os
from langsmith import traceable
from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate

# Set LangSmith API key (cloud-only, no self-hosting)
os.environ["LANGCHAIN_API_KEY"] = "your-langsmith-api-key"
os.environ["LANGCHAIN_TRACING_V2"] = "true"
os.environ["LANGCHAIN_ENDPOINT"] = "https://api.smith.langchain.com"

# Auto-tracing is enabled at module level
model = ChatOpenAI(model="gpt-4o-mini", temperature=0)
prompt = ChatPromptTemplate.from_template("Summarize this: {text}")
chain = prompt | model

# Chain automatically traced to LangSmith cloud
result = chain.invoke({"text": "Python is a language"})
print(result.content)

import os
from openai import OpenAI
from langfuse.openai import openai_integration
from langfuse import Langfuse

# Langfuse client (cloud or self-hosted via LANGFUSE_HOST)
os.environ["LANGFUSE_PUBLIC_KEY"] = "your-public-key"
os.environ["LANGFUSE_SECRET_KEY"] = "your-secret-key"
os.environ["LANGFUSE_HOST"] = "http://localhost:3000"  # Self-hosted, or omit for cloud

langfuse = Langfuse()
openai_integration(openai_integration, Langfuse())  # Enable Langfuse tracing for OpenAI SDK

# All OpenAI calls now traced to Langfuse (local or cloud)
client = OpenAI()
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Summarize Python"}]
)
print(response.choices[0].message.content)
langfuse.flush()  # Ensure traces are sent