Skip to main content

​
Overview

This guide shows you how to integrate Latitude Telemetry into an application that uses the Vercel AI SDK. The Vercel AI SDK can emit OpenTelemetry spans through its experimental_telemetry flag. Latitude can ingest those spans without a provider-specific instrumentation entry.
This guide covers Vercel AI SDK v6. On v7, OpenTelemetry moved into the separate @ai-sdk/otel package and became opt-out — see the Vercel AI SDK v7 guide instead.
You’ll keep calling the Vercel AI SDK exactly as you do today. Telemetry simply observes and enriches those calls.
The Vercel AI SDK integration is TypeScript only.

​
Requirements

  • A Latitude account and API key
  • A Latitude project slug
  • A Node.js project that uses the Vercel AI SDK (ai package)

​
Steps

1

Install

npm install @latitude-data/telemetry
2

Initialize and use

Initialize Latitude without an instrumentations array. The Vercel AI SDK emits its own OpenTelemetry spans. Pass latitude.getTracer("vercelai") to the AI SDK telemetry config on each call.
import { Latitude, capture } from "@latitude-data/telemetry"
import { generateText } from "ai"
import { openai } from "@ai-sdk/openai"

const latitude = new Latitude({
  apiKey: process.env.LATITUDE_API_KEY!,
  project: process.env.LATITUDE_PROJECT_SLUG!,
})

await capture("generate-support-reply", async () => {
  const { text } = await generateText({
    model: openai("gpt-4o"),
    prompt: "Hello",
    experimental_telemetry: {
      isEnabled: true,
      tracer: latitude.getTracer("vercelai"),
    },
  })
  return text
})

await latitude.shutdown()

​
Seeing Your Traces

Once connected, traces appear automatically in Latitude:
  1. Open your project in the Latitude dashboard
  2. Each execution shows input/output messages, model, token usage, latency, and errors