Vercel and OpenAI Integration

Vercel integrates with OpenAI to enable developers to build fast, scalable, and secure AI applications.

You can integrate with any OpenAI model through its REST API or JavaScript SDK, including the following OpenAI models:

• GPT-3.5: Understand and generate natural language or code
• GPT-4: An advanced model for natural language processing
• DALL·E: Generate and edit images from natural language
• Whisper: Convert audio to text
• Embeddings: Convert term into vectors

To help you get started, we have built a variety of AI templates integrating OpenAI with Vercel.

Vercel Postgres pgvector Starter

A Next.js template that uses Vercel Postgres as the database, pgvector for vector similarity search + OpenAI's text embedding models.

Next.js AI Chatbot

A full-featured, hackable Next.js AI chatbot built by Vercel

AI Film Recommendation Generator

Get curated show and movie recommendations with OpenAI GPT-3 API (text-davinci-003) and Vercel Edge Functions with streaming.

Before you begin, ensure you have an OpenAI account. Once registered:

Navigate to API Keys

Log into your OpenAI Dashboard and view API keys.

Generate API Key

Click on Create new secret key. Copy the generated API key securely.

Always keep your API keys confidential. Do not expose them in client-side code. Use Vercel Environment Variables for safe storage and do not commit these values to git.

Set Environment Variable

Finally, add the OPENAI_API_KEY environment variable in your project:

.env.local

OPENAI_API_KEY='sk-...3Yu5'

Integrating OpenAI into your Vercel project is seamless with the Vercel AI SDK.

Install the Vercel AI SDK in your project with your favorite package manager:

pnpm i ai

You can use the SDK to build AI applications with React (Next.js), Vue (Nuxt), Svelte (SvelteKit), and Solid (SolidStart).

The Vercel AI SDK also has experimental support for OpenAI Functions.

Learn more about using OpenAI Functions with the AI SDK.