Smolagents

2

Install smolagents and OpenAI

Required

Install smolagents and the OpenAI SDK. PostHog instruments your LLM calls by wrapping the OpenAI client, which you can pass to smolagents’ OpenAIServerModel.

pip install smolagents openai

3

Initialize PostHog and smolagents

Required

Initialize PostHog with your project token and host from your project settings (opens in a new tab), then create a PostHog OpenAI wrapper and pass it to smolagents’ OpenAIServerModel.

from smolagents import CodeAgent, OpenAIServerModel
from posthog.ai.openai import OpenAI
from posthog import Posthog
posthog = Posthog(
    "<ph_project_token>",
    host="https://us.i.posthog.com"
)
openai_client = OpenAI(
    api_key="your_openai_api_key",
    posthog_client=posthog
)
model = OpenAIServerModel(
    model_id="gpt-5-mini",
    client=openai_client,
)

How this works

PostHog’s OpenAI wrapper is a drop-in replacement for openai.OpenAI. By passing it as the client to OpenAIServerModel, all LLM calls made by smolagents are automatically captured as $ai_generation events.

4

Run your agent

Required

Use smolagents as normal. PostHog automatically captures an $ai_generation event for each LLM call made through the wrapped OpenAI client.

agent = CodeAgent(
    tools=[],
    model=model,
)
result = agent.run(
    "What is a fun fact about hedgehogs?"
)
print(result)

You can expect captured $ai_generation events to have the following properties:

Verify traces and generations

Recommended

Confirm LLM events are being sent to PostHog

Let’s make sure LLM events are being captured and sent to PostHog. Under LLM analytics, you should see rows of data appear in the Traces and Generations tabs.

Check for LLM events in PostHog (opens in a new tab)

5

Next steps

Recommended

Now that you’re capturing AI conversations, continue with the resources below to learn what else LLM Analytics enables within the PostHog platform.