Trace with PydanticAI

LangSmith can capture traces generated by PydanticAI using its built-in OpenTelemetry instrumentation. This guide shows you how to automatically capture traces from your PydanticAI agents and send them to LangSmith for monitoring and analysis.

Installation

Install the required packages:

pip

pip install langsmith pydantic-ai opentelemetry-exporter-otlp

uv add langsmith pydantic-ai opentelemetry-exporter-otlp

Requires LangSmith Python SDK version langsmith>=0.4.26 for optimal OpenTelemetry support.

Setup

1. Configure environment variables

Set your API keys and project name:

export LANGSMITH_API_KEY=<your_langsmith_api_key>
export LANGSMITH_PROJECT=<your_project_name>
export OPENAI_API_KEY=<your_openai_api_key>

2. Configure OpenTelemetry integration

In your PydanticAI application, configure the LangSmith OpenTelemetry integration:

from langsmith.integrations.otel import configure
from pydantic_ai import Agent

# Configure LangSmith tracing
configure(project_name="pydantic-ai-demo")

# Instrument all PydanticAI agents
Agent.instrument_all()

You do not need to set any OpenTelemetry environment variables or configure exporters manually—configure() handles everything automatically.

3. Create and run your PydanticAI agent

Once configured, your PydanticAI agents will automatically send traces to LangSmith:

from langsmith.integrations.otel import configure
from pydantic_ai import Agent

# Configure LangSmith tracing
configure(project_name="pydantic-ai-demo")

# Instrument all PydanticAI agents
Agent.instrument_all()

# Create and run an agent
agent = Agent('openai:gpt-4o')
result = agent.run_sync('What is the capital of France?')
print(result.output)
#> Paris

Advanced usage

Custom metadata and tags

You can add custom metadata to your traces using OpenTelemetry span attributes:

from opentelemetry import trace
from pydantic_ai import Agent
from langsmith.integrations.otel import configure

configure(project_name="pydantic-ai-metadata")
Agent.instrument_all()

tracer = trace.get_tracer(__name__)

agent = Agent('openai:gpt-4o')

with tracer.start_as_current_span("pydantic_ai_workflow") as span:
    span.set_attribute("langsmith.metadata.user_id", "user_123")
    span.set_attribute("langsmith.metadata.workflow_type", "question_answering")
    span.set_attribute("langsmith.span.tags", "pydantic-ai,production")

    result = agent.run_sync('Explain quantum computing in simple terms')
    print(result.output)

Combine with other instrumentors

You can combine PydanticAI instrumentation with other OpenTelemetry instrumentors:

from langsmith.integrations.otel import configure
from pydantic_ai import Agent
from openinference.instrumentation.openai import OpenAIInstrumentor

# Configure LangSmith tracing
configure(project_name="multi-framework-app")

# Initialize multiple instrumentors
Agent.instrument_all()
OpenAIInstrumentor().instrument()

# Your application code using multiple frameworks

---

Edit the source of this page on GitHub.

Connect these docs programmatically to Claude, VSCode, and more via MCP for real-time answers.