Vercel AI SDK

Use Cortecs with the Vercel AI SDK

The Vercel AI SDK supports both the Responses API and OpenAI-compatible Chat Completions. Use Responses for new agent workflows.

Responses API

Install the required packages:

npm install ai @ai-sdk/openai

Configure the OpenAI provider with the Cortecs base URL:

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

const cortecs = createOpenAI({
  name: "cortecs",
  apiKey: process.env.CORTECS_API_KEY,
  baseURL: "https://api.cortecs.ai/v1",
});

const { text } = await generateText({
  model: cortecs.responses(process.env.CORTECS_MODEL!),
  prompt: "Explain model routing in two sentences.",
});

console.log(text);

Choose a model that advertises the capabilities your application uses in the Models API.

Chat Completions fallback

If a workflow or middleware expects Chat Completions, use the OpenAI-compatible provider:

npm install @ai-sdk/openai-compatible

import { createOpenAICompatible } from "@ai-sdk/openai-compatible";
import { generateText } from "ai";

const cortecs = createOpenAICompatible({
  name: "cortecs",
  apiKey: process.env.CORTECS_API_KEY,
  baseURL: "https://api.cortecs.ai/v1",
});

const { text } = await generateText({
  model: cortecs.chatModel(process.env.CORTECS_MODEL!),
  prompt: "Explain model routing in two sentences.",
  providerOptions: {
    cortecs: {
      preference: "balanced",
    },
  },
});

console.log(text);

Provider options are forwarded in the request body by the OpenAI-compatible provider. See Advanced Usage for the Cortecs routing fields you can pass.

See the Vercel AI SDK documentation for the OpenAI provider and OpenAI-compatible providers.