Vercel AI SDK observability installation

Let AI instrument your LLM calls for you

Skip the manual setup — run this in your project and the wizard installs the SDK and wires up AI Observability for you.

Learn more
PostHog Wizard hedgehog

Contents

  1. Install dependencies

    Required

    Install the PostHog AI package, the Vercel AI SDK, the OpenTelemetry SDK, and Zod for defining tool schemas.

    npm install @posthog/ai "ai@^6" "@ai-sdk/openai@^3" @opentelemetry/sdk-node @opentelemetry/resources zod

    AI SDK version: this integration requires AI SDK v5 or v6. AI SDK v7 removed its OpenTelemetry instrumentation, so it no longer emits the ai.* spans that PostHogSpanProcessor reads, and telemetry.metadata is no longer accepted. Installing ai without a version constraint gets you v7, which will not produce any events. Pin ai@^6 with the matching @ai-sdk/openai@^3 until PostHog ships v7 support.

  2. Set up the OpenTelemetry exporter

    Required

    Initialize the OpenTelemetry SDK with PostHog's PostHogSpanProcessor. This sends gen_ai.* spans directly to PostHog's OTLP ingestion endpoint. PostHog converts these into $ai_generation events automatically.

    import { NodeSDK } from '@opentelemetry/sdk-node'
    import { resourceFromAttributes } from '@opentelemetry/resources'
    import { PostHogSpanProcessor } from '@posthog/ai/otel'
    

    const sdk = new NodeSDK({
      resource: resourceFromAttributes({
        'service.name': 'my-app',
      }),
      spanProcessors: [
        new PostHogSpanProcessor({
          projectToken: '<ph_project_token>',
          host: 'https://us.i.posthog.com',
        }),
      ],
    })
    sdk.start()

  3. Call Vercel AI with telemetry enabled

    Required

    Pass experimental_telemetry to your Vercel AI SDK calls. The posthog_distinct_id metadata field links events to a specific user in PostHog. Define tools the same way you always would, with an execute function, as get_weather does below.

    import { generateText, tool, stepCountIs } from 'ai'
    import { openai } from '@ai-sdk/openai'
    import { z } from 'zod'
    

    const result = await generateText({
      model: openai('gpt-5-mini'),
      prompt: "What's the weather in Paris?",
      tools: {
        get_weather: tool({
          description: 'Get the weather for a city',
          inputSchema: z.object({ city: z.string() }),
          execute: async ({ city }) => `It's always sunny in ${city}!`,
        }),
      },
      stopWhen: stepCountIs(5), // let the model see the tool result and respond
      experimental_telemetry: {
        isEnabled: true,
        functionId: 'my-ai-function',
        metadata: {
          posthog_distinct_id: 'user_123', // optional
          posthog_environment: 'production', // custom property: sets "environment" on the event
          $ai_session_id: 'conversation-abc', // optional: groups calls into one session
        },
      },
    })
    

    console.log(result.text)

    Note: If you want to capture LLM events anonymously, omit the posthog_distinct_id metadata field. See our docs on anonymous vs identified events to learn more.

    Custom properties: Prefix any telemetry metadata field with posthog_ to attach it to the $ai_generation event as a custom property. PostHog strips the prefix, so posthog_environment becomes an environment property you can filter and break down by. PostHog does not capture other metadata fields.

    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

  4. 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 AI Observability, you should see rows of data appear in the Traces and Generations tabs.


    LLM generations in PostHog
    Check for LLM events in PostHog

  5. Next steps

    Recommended

    Now that you're capturing AI conversations, continue with the resources below to learn what else AI Observability 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?