Vercel AI SDK

​

Initializing HoneyHive Tracer

Use the following code to initialize HoneyHive tracing in your project:

• TypeScript

import { HoneyHiveTracer } from "honeyhive";

const tracer = await HoneyHiveTracer.init({
  apiKey: process.env.HH_API_KEY,
  project: process.env.HH_PROJECT,
  sessionName: "your-session-name",
});

​

Supported Vercel AI SDK Versions/Interfaces

Compatible with Vercel AI SDK versions supporting the experimental telemetry feature. Please check the Vercel AI SDK documentation for specific version compatibility.

For the most up-to-date compatibility information, please refer to the HoneyHive documentation.

​

Integrating with generateText

The Vercel AI SDK’s generateText function supports telemetry integration through the experimental_telemetry option. Here’s how to enable tracing:

const result = await generateText({
  model: openai('your-model-name'),
  prompt: 'Your prompt here',
  experimental_telemetry: {
    isEnabled: true,
    tracer: tracer.getTracer(),
  },
});

​

Nesting

Nesting is handled automatically by the HoneyHive tracing system. When you use traced components within other traced components, the system will create a hierarchical structure of spans, reflecting the nested nature of your Vercel AI SDK operations.

​

Enriching Properties

For information on how to enrich your traces and spans with additional context, see our enrichment documentation.

​

Adding Evaluators

Once traces have been logged in the HoneyHive platform, you can then run evaluations with either Python or TypeScript.

​

Cookbook Examples

​

TypeScript Example

import { generateText } from 'ai';
import { openai } from '@ai-sdk/openai';
import { WrapperTracer, HoneyHiveTracer } from "honeyhive";

const HH_API_KEY = process.env.HH_API_KEY || "";
const HH_API_URL = process.env.HH_API_URL;
const HH_PROJECT = process.env.HH_PROJECT || "";
const HH_PROJECT_ID = process.env.HH_PROJECT_ID || "";

async function initializeTracer(sessionName: string): Promise<HoneyHiveTracer> {
  const tracer = await HoneyHiveTracer.init({
    apiKey: HH_API_KEY,
    project: HH_PROJECT,
    sessionName: sessionName,
  });

  return tracer;
}

async function main() {
  const tracer = await initializeTracer('vercelTest');
  const result = await generateText({
    model: openai('gpt-4-turbo'),
    prompt: 'Write a short story about a cat.',
    experimental_telemetry: {
      isEnabled: true,
      tracer: tracer.getTracer(),
    },
  });

  console.log(result.text);
}

main().catch(console.error);

This example demonstrates how to integrate HoneyHive tracing with the Vercel AI SDK in a TypeScript environment, covering tracer initialization and text generation with OpenAI models.