Connect Vercel AI SDK in 1 line.
Get wholesale edge caching automatically.

npm install ai @ai-sdk/openai

import { createOpenAI } from '@ai-sdk/openai';
import { generateText } from 'ai';

// Initialize client pointing to Newton's Edge Gateway
const newtons = createOpenAI({
  baseURL: 'https://newtons.dev/api/v1',
  apiKey: process.env.NEWTONS_API_KEY,
});

export async function POST(req: Request) {
  const { prompt } = await req.json();

  const { text, usage } = await generateText({
    model: newtons('claude-sonnet-5-5'), // or 'claude-opus-5-5', 'claude-haiku-4-5-20251001'
    prompt,
    // Prompt caching and edge routing happen automatically
  });

  return Response.json({ text, usage });
}

Standard Wire Protocol

No custom SDK wrappers or vendor lock-in. Keep using official OpenAI and Anthropic client libraries by just modifying your `baseURL`.

85% Prompt Discounts

Prompt caching discounts from Anthropic, OpenAI, and DeepSeek are passed through directly at wholesale rates without gatekeeper markups.

One API Key for Everything

Eliminate managing multiple enterprise invoices and prepaid balances across separate AI model providers.