OpenAI API Issues Integration to JS

I’m currently testing integrating OpenAI’s GPT API on a small-scale project that involves index.html(web-page), index.cjs (where the code is run and was previously index.js) And then the necessary files such as the .env , package-lock.json, node_modules , and package.json. I’ve done the following commands via gitbash already:
-npm install dotenv
-npm install openai

I declared dotenv at the beginning of my index.cjs file and I correctly set up my API key in the .env by the following format: OPENAI_API_KEY = example_api

I did the following troubleshooting steps:

1. verified .env file is present in under the root project directory and that the API key is correctly configured.
2. Attempted to print the API key to the console but got “undefined”.
   3.Attempted different API keys from my account and a friends whom we both have no usage and are on a free plan. I used the gpt-3.5-turbo model, somehow I got error 429 on both given that neither has ever used it.
   4.Restarted VS code to make sure changes to the .env file and environment variable were applied.
   5.Restarted GitBash.
3. Got the following errors: “RateLimitError: 429″(Exceeding quota), “NotFoundError: 404″(Model unavailable when I had used GPT-4”, and “AuthenticationError”(incorrect API key).

What I Tried:
I checked my .env file to ensure the OPENAI_API_KEY is correctly set and loaded using the dotenv package.
I logged the API key to the console to verify it was being loaded correctly.
I tested the connection to OpenAI’s API by running a script that attempts to generate a haiku using the gpt-3.5-turbo model.
I also tried using different API keys and different OpenAI models to see if they work under the free plan.
I checked my OpenAI dashboard to confirm my quota usage was at 0 and that I was on the free plan with no cost.
What I Expected:
I expected the OpenAI API key to load correctly from the .env file.
I anticipated the haiku generation to succeed, without hitting rate limits or quota errors.
I thought that the gpt-3.5-turbo model would work under the free tier without exceeding usage limits, given that no cost had been incurred yet.
I expected to see the generated haiku printed in the console if everything was set up correctly.
What Actually Happened:
The API key loaded correctly, as confirmed by the logged output.
However, despite having no usage cost, I received an “insufficient quota” error (status 429) indicating that I had exceeded my current quota, even though no usage had occurred.
The model I selected (gpt-3.5-turbo) should be available on the free plan, but I encountered a rate limit error suggesting a quota issue.
No haiku was generated due to this error, and instead, the error message pointed to an issue with my plan or quota. The following code:

index.cjs:

// Load environment variables before anything else
require('dotenv').config();


// Import the OpenAI client
const { OpenAI } = require('openai');

// Initialize the OpenAI client with the API key from the environment
const openai = new OpenAI({
    apiKey: process.env.OPENAI_API_KEY,  // The API key is automatically read from the .env file
});

async function generateHaiku() {
    try {
        // Call OpenAI's API to generate a haiku based on a prompt
        const completion = await openai.chat.completions.create({
            model: "gpt-3.5-turbo", // Using GPT-3.5 for the free plan
            messages: [
                { role: "system", content: "You are a helpful assistant." },
                {
                    role: "user",
                    content: "Write a haiku about recursion in programming.",
                },
            ],
        });

        // Log the generated haiku
        console.log("Generated Haiku:", completion.choices[0].message.content);
    } catch (error) {
        // Catch any errors and log them
        console.error("Error generating haiku:", error);
    }
}

// Call the function to generate the haiku
generateHaiku();