Netlify AI Gateway

Netlify AI Gateway provides access to AI models from multiple providers without managing API keys directly. It is available on all Netlify sites.

How It Works

The AI Gateway acts as a proxy — you use standard provider SDKs (OpenAI, Anthropic, Google) but point them at Netlify's gateway URL instead of the provider's API. Netlify handles authentication, rate limiting, and monitoring.

Setup

1. Enable AI on your site in the Netlify UI
2. The environment variable OPENAI_BASE_URL is set automatically by Netlify
3. Install the provider SDK you want to use

No provider API keys are needed — Netlify's gateway handles authentication.

Using OpenAI SDK

npm install openai

import OpenAI from "openai";

const openai = new OpenAI();
// OPENAI_BASE_URL is auto-configured — no API key or base URL needed

const completion = await openai.chat.completions.create({
  model: "gpt-4o-mini",
  messages: [{ role: "user", content: "Hello!" }],
});

Using Anthropic SDK

npm install @anthropic-ai/sdk

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  baseURL: Netlify.env.get("ANTHROPIC_BASE_URL"),
});

const message = await client.messages.create({
  model: "claude-sonnet-4-5-20250929",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello!" }],
});

Using Google AI SDK

npm install @google/generative-ai

import { GoogleGenerativeAI } from "@google/generative-ai";

const genAI = new GoogleGenerativeAI("placeholder");
// Configure base URL via environment variable

const model = genAI.getGenerativeModel({ model: "gemini-2.5-flash" });
const result = await model.generateContent("Hello!");

In a Netlify Function

import type { Config, Context } from "@netlify/functions";
import OpenAI from "openai";

export default async (req: Request, context: Context) => {
  const { prompt } = await req.json();
  const openai = new OpenAI();

  const completion = await openai.chat.completions.create({
    model: "gpt-4o-mini",
    messages: [{ role: "user", content: prompt }],
  });

  return Response.json({
    response: completion.choices[0].message.content,
  });
};

export const config: Config = {
  path: "/api/ai",
  method: "POST",
};

Environment Variables

These are configured automatically when AI is enabled on the site. No manual setup required.

Local Development

With @netlify/vite-plugin or netlify dev, gateway environment variables are injected automatically. The AI Gateway is accessible during local development after the site has been deployed at least once.

Available Models

The gateway supports models from OpenAI, Anthropic, and Google. Use standard model identifiers (e.g., gpt-4o-mini, claude-sonnet-4-5-20250929, gemini-2.5-flash). Available models may change — refer to Netlify's AI Gateway documentation for the current list.