Thank you for this plugin, i just have a question,

how to use this plugin with gpt-4 ?

ChatGPT has the "stream" way to respond to the answer, how to do that?

While routing all of the requests through the nitro server does a great job of keeping our API keys private, it also has a potentially unintended effect for some users.

If your intended Chat GPT user interface component is behind authentication (because who wants to expose unlimited use of your api key to the public?) the routes that are exposed with this module are always public.

I could be missing some intended configuration for this use-case, but as far as I can tell the routes /api/chat and /api/chat-completion are always exposed to the internet. Again, this could be intended functionality, I just wanted to make sure that this was not a bug. For my particular uses, I will have to remove this module for this reason.

Thank you for the great work and contribution nonetheless!

The docs state:

// entirely optional
  chatgpt: {
    apiKey: 'Your apiKey here goes here'
  },

Wouldn't you want to reference the key within the runtimeConfig property like so?

	runtimeConfig: {
		OPENAI_API_KEY: "" //<---- would be copied form an `.env` file
	},

Hi,
your project is awesome and it works great with almost no effort :)

Nevertheless I would suggest to change the interface of the chatCompletion function:

const chatCompletion = async (message: IMessage, model?: IModel, options?: IOptions)

to

const chatCompletion = async (messages: IMessages, model?: IModel, options?: IOptions)

or something similar :)

According to openAI API specification, chatCompletions can be configured with a set of messages:

const response = await openai.chat.completions.create({
model: "gpt-3.5-turbo",
messages: [{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Who won the world series in 2020?"},
{"role": "assistant", "content": "The Los Angeles Dodgers won the World Series in 2020."},
{"role": "user", "content": "Where was it played?"}],
});

This change would allow to implement easily use cases such as the "Socratic Tutor" https://platform.openai.com/examples/default-socratic-tutor.

All the best,
Mauro

This looks amazing! A couple of points to mention, in case you find it helpful:

1. You do not need to stringify the body of $fetch requests, or set content-type. This will be done automatically.

2. Because you set the public key of runtimeConfig, the user's API key is exposed to the client.

   ⚠️ This should probably be fixed as a matter of urgency.

   https://github.com/SchnapsterDog/nuxt-chatgpt/blob/master/src/module.ts#L50C5-L52. (You just need to update that same line to remove public.)

Expected Behavior

• A response related to the prompt the user provides.

Actual Behavior

• A totally random response.

Steps to Reproduce the Problem

Basically use the library by making a simple request according to their example.

Specifications

• Version: 0.1.8
• Platform: MacOS
• Subsystem: Unix

Im not sure why iam getting this error do i need to pay or the free trial can be used?