Vercel AI SDK · Laras Router Docs

This guide connects Vercel AI SDK to Laras Router. Every step was checked against the Vercel AI SDK documentation.

You need an API key (lr_live_…) and a topped-up balance. New to Laras Router? Start with the Quickstart.

Set up

Install the packages

Install the AI SDK and the OpenAI-compatible provider.

Shell

npm install ai @ai-sdk/openai-compatible

Set your API key

Put your key in an environment variable.

Shell

export LARAS_API_KEY="lr_live_your_key_here"

Generate text

Create the provider and generate text.

TypeScript

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

const laras = createOpenAICompatible({
  name: "larasrouter",
  baseURL: "https://api.larasrouter.com/v1",
  apiKey: process.env.LARAS_API_KEY,
});

const { text } = await generateText({
  model: laras("deepseek-flash"),
  prompt: "Hello!",
});
console.log(text);

Good to know

Use @ai-sdk/openai-compatible. Since AI SDK 5, the @ai-sdk/openai provider calls the Responses API by default.
For streaming, use streamText. To receive token usage while streaming, set includeUsage: true when you create the provider.

Settings and screens can change between versions. If something looks different, check the Vercel AI SDK documentation. For other tools, see all integrations.