AutoGen LLM analytics installation

  1. Install the PostHog SDK

    Required

    Setting up analytics starts with installing the PostHog SDK. The AutoGen integration uses PostHog's OpenAI wrapper since AutoGen uses OpenAI under the hood.

    pip install posthog

  2. Install AutoGen

    Required

    Install AutoGen with the OpenAI extension. PostHog instruments your LLM calls by wrapping the OpenAI client that AutoGen uses internally.

    pip install "autogen-agentchat" "autogen-ext[openai]"

  3. Initialize PostHog and AutoGen

    Required

    Initialize PostHog with your project API key and host from your project settings, then create a PostHog OpenAI wrapper and pass it to AutoGen's OpenAIChatCompletionClient.

    import asyncio
    from posthog.ai.openai import OpenAI
    from posthog import Posthog
    from autogen_agentchat.agents import AssistantAgent
    from autogen_ext.models.openai import OpenAIChatCompletionClient
    

    posthog = Posthog(
        "<ph_project_api_key>",
        host="https://us.i.posthog.com"
    )
    

    openai_client = OpenAI(
        api_key="your_openai_api_key",
        posthog_client=posthog,
    )
    

    model_client = OpenAIChatCompletionClient(
        model="gpt-4o",
        openai_client=openai_client,
    )
    How this works

    AutoGen's OpenAIChatCompletionClient accepts a custom OpenAI client via the openai_client parameter. PostHog's OpenAI wrapper is a proper subclass of openai.OpenAI, so it works directly. PostHog captures $ai_generation events automatically without proxying your calls.

  4. Run your agents

    Required

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

    agent = AssistantAgent("assistant", model_client=model_client)
    

    async def main():
        result = await agent.run(task="Say 'Hello World!'")
        print(result)
        await model_client.close()
    

    asyncio.run(main())

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

    PropertyDescription
    $ai_modelThe specific model, like gpt-5-mini or claude-4-sonnet
    $ai_latencyThe latency of the LLM call in seconds
    $ai_time_to_first_tokenTime to first token in seconds (streaming only)
    $ai_toolsTools and functions available to the LLM
    $ai_inputList of messages sent to the LLM
    $ai_input_tokensThe number of tokens in the input (often found in response.usage)
    $ai_output_choicesList of response choices from the LLM
    $ai_output_tokensThe number of tokens in the output (often found in response.usage)
    $ai_total_cost_usdThe total cost in USD (input + output)
    [...]See full list of properties

  5. 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.


    LLM generations in PostHog
    Check for LLM events in PostHog

  6. 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.

    ResourceDescription
    BasicsLearn the basics of how LLM calls become events in PostHog.
    GenerationsRead about the $ai_generation event and its properties.
    TracesExplore the trace hierarchy and how to use it to debug LLM calls.
    SpansReview spans and their role in representing individual operations.
    Anaylze LLM performanceLearn how to create dashboards to analyze LLM performance.

Community questions

Was this page useful?

Questions about this page? or post a community question.