Vercel AI SDK

Learn how to integrate Respan tracing with the Vercel AI SDK (Next.js) using OpenTelemetry.

What is Vercel AI SDK tracing?

This guide shows how to set up Respan tracing with Next.js and the Vercel AI SDK so you can monitor and trace your AI-powered applications.

Steps to use

If you already have a Next.js + Vercel AI SDK app, start from Step 1 below.

1. Install Respan exporter

   Install the Respan exporter package:

   npm

   npm install @respan/exporter-vercel
   
2. Set up OpenTelemetry instrumentation

   Next.js supports OpenTelemetry instrumentation out of the box. Create instrumentation.ts in your project root (where package.json lives).

   Install Vercel’s OpenTelemetry instrumentation:

   yarn add @vercel/otel
   

   Then configure the Respan exporter:

   instrumentation.ts

   import { registerOTel } from "@vercel/otel";
   import { RespanExporter } from "@respan/exporter-vercel";
   
   export function register() {
     registerOTel({
       serviceName: "next-app",
       traceExporter: new RespanExporter({
         apiKey: process.env.RESPAN_API_KEY,
         baseUrl: process.env.RESPAN_BASE_URL,
         debug: true,
       }),
     });
   }
   
3. Configure environment variables

   Add your Respan credentials (and your provider key) to .env.local:

   OpenAI

   .env.local

   OPENAI_API_KEY=your_openai_api_key_here
   
   RESPAN_API_KEY=your_respan_api_key_here
   RESPAN_BASE_URL=https://api.respan.ai
   

   Anthropic

   .env.local

   ANTHROPIC_API_KEY=your_anthropic_api_key_here
   
   RESPAN_API_KEY=your_respan_api_key_here
   RESPAN_BASE_URL=https://api.respan.ai
   

   Google Gemini

   .env.local

   GOOGLE_GENERATIVE_AI_API_KEY=your_google_api_key_here
   
   RESPAN_API_KEY=your_respan_api_key_here
   RESPAN_BASE_URL=https://api.respan.ai
   
4. Enable telemetry in your route

   In your API route file (e.g. app/api/chat/route.ts), enable telemetry by adding the experimental_telemetry option.

   OpenAI

   app/api/chat/route.ts

   import { openai } from "@ai-sdk/openai";
   import { streamText } from "ai";
   
   export const maxDuration = 30;
   
   export async function POST(req: Request) {
     const { messages, id } = await req.json();
     console.log("chat id", id);
   
     const result = streamText({
       model: openai("gpt-4o"),
       messages,
       experimental_telemetry: {
         isEnabled: true,
         metadata: {
           customer_identifier: "customer_from_metadata",
         },
       },
     });
   
     return result.toDataStreamResponse();
   }
   

   Anthropic

   app/api/chat/route.ts

   import { anthropic } from "@ai-sdk/anthropic";
   import { streamText } from "ai";
   
   export const maxDuration = 30;
   
   export async function POST(req: Request) {
     const { messages } = await req.json();
   
     const result = streamText({
       model: anthropic("claude-3-5-sonnet-20240620"),
       messages,
       experimental_telemetry: {
         isEnabled: true,
       },
     });
   
     return result.toDataStreamResponse();
   }
   

   Google Gemini

   app/api/chat/route.ts

   import { google } from "@ai-sdk/google";
   import { streamText } from "ai";
   
   export const maxDuration = 30;
   
   export async function POST(req: Request) {
     const { messages } = await req.json();
   
     const result = streamText({
       model: google("gemini-2.0-flash"),
       messages,
       experimental_telemetry: {
         isEnabled: true,
       },
     });
   
     return result.toDataStreamResponse();
   }
   
5. Run locally and verify traces

   1. Start your dev server:

   pnpm

   pnpm dev
   

   2. Make some chat requests through your application
   3. Verify traces in Respan:
      • Go to Logs → Traces in your Respan dashboard
      • Confirm requests are being traced

   If you hit missing dependency errors from @vercel/otel, install the missing packages and retry.

What gets traced

With this setup, Respan will capture:

• AI model calls: requests made via the Vercel AI SDK
• Token usage: input and output token counts
• Performance metrics: latency and throughput
• Errors: failed requests and error details
• Custom metadata: additional context you attach via telemetry metadata