Vercel AI SDK Integration

Route all Vercel AI SDK calls through the Curate-Me gateway for automatic cost tracking, rate limiting, and PII scanning. The SDK’s createOpenAI provider accepts a custom baseURL — one property changed.

Before and after


// Before — direct to OpenAI
import { createOpenAI } from '@ai-sdk/openai';
import { generateText } from 'ai';
 
const openai = createOpenAI({
  apiKey: 'sk-your-openai-key',
});
 
// After — through Curate-Me gateway
import { createOpenAI } from '@ai-sdk/openai';
import { generateText } from 'ai';
 
const openai = createOpenAI({
  apiKey: 'sk-your-openai-key',
  baseURL: 'https://api.curate-me.ai/v1/openai',        // ← added
  headers: { 'X-CM-API-Key': 'cm_sk_YOUR_KEY' },        // ← added
});

Every generateText, streamText, and generateObject call that uses this provider is now governed by the gateway.

Full example


import { createOpenAI } from '@ai-sdk/openai';
import { generateText, streamText } from 'ai';
 
const openai = createOpenAI({
  apiKey: 'sk-your-openai-key',
  baseURL: 'https://api.curate-me.ai/v1/openai',
  headers: {
    'X-CM-API-Key': 'cm_sk_YOUR_KEY',
    'X-CM-Tags': 'project=vercel-ai-app,env=production',  // optional cost tags
  },
});
 
// Non-streaming
const { text, usage } = await generateText({
  model: openai('gpt-4o-mini'),
  prompt: 'What is AI governance?',
});
console.log(text);
console.log(`Tokens: prompt=${usage.promptTokens}, completion=${usage.completionTokens}`);
 
// Streaming
const stream = streamText({
  model: openai('gpt-4o-mini'),
  prompt: 'List three benefits of LLM cost tracking.',
});
for await (const chunk of stream.textStream) {
  process.stdout.write(chunk);
}
 
// With system prompt and messages
const { text: answer } = await generateText({
  model: openai('gpt-4o-mini'),
  system: 'You are a concise technical advisor.',
  messages: [
    { role: 'user', content: 'How does rate limiting protect AI apps?' },
  ],
});
console.log(answer);

Next.js route handler


// app/api/chat/route.ts
import { createOpenAI } from '@ai-sdk/openai';
import { streamText } from 'ai';
 
const openai = createOpenAI({
  apiKey: process.env.OPENAI_API_KEY!,
  baseURL: 'https://api.curate-me.ai/v1/openai',
  headers: { 'X-CM-API-Key': process.env.CM_API_KEY! },
});
 
export async function POST(req: Request) {
  const { messages } = await req.json();
 
  const result = streamText({
    model: openai('gpt-4o-mini'),
    messages,
  });
 
  return result.toDataStreamResponse();
}

Prerequisites


npm install ai @ai-sdk/openai

What you get

Every Vercel AI SDK call through the gateway automatically receives:

Feature	Description
Cost tracking	Per-request token and dollar cost recorded in real time
Rate limiting	Per-org, per-key request throttling
PII scanning	Regex scan for secrets and PII before hitting the provider
Model allowlists	Only approved models per your org policy
Budget caps	Daily and monthly spend limits enforced
Audit trail	Full request metadata logged for compliance

Working example

TypeScript example — generateText, streamText, system prompts

Next steps

Gateway Quickstart — full setup walkthrough
OpenAI SDK Integration — Python and TypeScript examples
Cost Tracking — budget alerts and cost attribution