Bookworm

How it works

This app leverages OpenAI's recently released ChatGPT API with gpt-3.5-turbo model to respond to a chain of chat messages. Users submit messages to a SvelteKit API Endpoint/Request Handler, which relays the messages to the ChatGPT API. The responses are then proxied back to the client via SSE to stream the response in realtime.

Built with

• Meta-Framework: SvelteKit
• Styles/Components: TailwindCSS & DaisyUI
• Deployment: Vercel

Run Locally

Clone the repository

git clone https://github.com/huntabyte/chatty

Create a .env file within the new directory

cd chatty && touch .env
echo OPENAI_KEY=<YOUR_API_KEY_HERE> >> .env

Install dependencies & start the dev server

pnpm i && pnpm run dev

You can now access the dev server running at localhost:5173

Deploy to Vercel

Commit the repository to GitHub and select it when creating a new Vercel deployment.

Don't forget to set the OPENAI_KEY environment variable within your Vercel project settings.