Skip to main content

Quickstart

This Quickstart should help new developers get started quickly with ragged, leveraging its features for efficient LLM integrations.

Installation​

You'll need to install ragged along with its peer dependencies, openai for LLM driver support (we only support openai right now) and rxjs for streaming.

# npm
npm install --save --save-exact openai rxjs ragged

#pnpm
pnpm install --save --save-exact openai rxjs ragged

#yarn
yarn add -E openai rxjs ragged

Your first Ragged LLM call​

import { Ragged } from "ragged";

const OPENAI_API_KEY = "your api key"

const r = new Ragged({
  provider: "openai",
  config: {
    apiKey: OPENAI_API_KEY,
    // You need the following line if you're in a browser. See OpenAI client docs.
    dangerouslyAllowBrowser: true
  },
});

r.chat("What is Toronto?")
  .firstText() // convert to promise
  .then(console.log) // Toronto is a city in Canada. It has a population of...

Explore the documentation​

You're now all set to start building really cool LLM apps! Ragged features very easy and fluent APIs for streaming, tool use, and more. You can see More Examples here for inspiration!