Skip to main content

Get started with LangSmith

LangSmith is a platform for building production-grade LLM applications. It allows you to closely monitor and evaluate your application, so you can ship quickly and with confidence. Use of LangChain is not necessary - LangSmith works on its own!

1. Install LangSmith

pip install -U langsmith

2. Create an API key

To create an API key head to the Settings page. Then click Create API Key.

3. Set up your environment

export LANGCHAIN_API_KEY=<your-api-key>

# The below examples use the OpenAI API, though it's not necessary in general
export OPENAI_API_KEY=<your-openai-api-key>

4. Log your first trace

We provide multiple ways to log traces to LangSmith. Below, we'll highlight how to use traceable. See more on the Annotate code for tracing page.

import openai
from langsmith.wrappers import wrap_openai
from langsmith import traceable

# Auto-trace LLM calls in-context
client = wrap_openai(openai.Client())

@traceable # Auto-trace this function
def pipeline(user_input: str):
result =
messages=[{"role": "user", "content": user_input}],
return result.choices[0].message.content

pipeline("Hello, world!")
# Out: Hello there! How can I assist you today?

5. Run your first evaluation

Evaluation requires a system to test, data to serve as test cases, and optionally evaluators to grade the results. Here we use a built-in accuracy evaluator.

from langsmith import Client
from langsmith.evaluation import evaluate

client = Client()

# Define dataset: these are your test cases
dataset_name = "Sample Dataset"
dataset = client.create_dataset(dataset_name, description="A sample dataset in LangSmith.")
{"postfix": "to LangSmith"},
{"postfix": "to Evaluations in LangSmith"},
{"output": "Welcome to LangSmith"},
{"output": "Welcome to Evaluations in LangSmith"},

# Define your evaluator
def exact_match(run, example):
return {"score": run.outputs["output"] == example.outputs["output"]}

experiment_results = evaluate(
lambda input: "Welcome " + input['postfix'], # Your AI system goes here
data=dataset_name, # The data to predict and grade over
evaluators=[exact_match], # The evaluators to score the results
experiment_prefix="sample-experiment", # The name of the experiment
"version": "1.0.0",
"revision_id": "beta"

Was this page helpful?

You can leave detailed feedback on GitHub.