LLM API Access

Updated: 2025-11-24

To programmatically interact with LLMs, you need to set up API access. Follow these steps to obtain your GPT API key and configure it securely in PyCharm.

While we use OpenAI GPT APIs for this course, the interaction methods are similar across other LLMs such as Anthropic Claude and Google Gemini.

OpenAI Account

Create an OpenAI account (if needed):

• Visit platform.openai.com/signup

• Sign up with your email address or use an existing account

Generate your API key:

• Log in to https://platform.openai.com

• Visit platform.openai.com/api-keys

• Click "+ Create new secret key"

• Give your key a descriptive name (e.g., "NLP-Essentials")

• Click "Create secret key"

• Copy the key immediately and save it securely - you will not be able to see it again!

Add Billing Information (if required):

• OpenAI requires payment information to use the API

• Navigate to Settings > Billing > Payment methods

• Add a payment method

• Consider setting usage limits to prevent unexpected charges

• New accounts may receive free trial credits; check your usage page to see your balance

Security Warning: Never share your API key or commit it to GitHub. Treat it like a password. If you accidentally expose it, delete it immediately from the OpenAI dashboard and create a new one.

Configure API Key

Install the python-dotenv package:

pip install python-dotenv

Create a .env file in your project root directory (nlp-essentials/) and add your OpenAI API Key by replacing your-api-key-here below:

OPENAI_API_KEY=your-api-key-here

Add .env to your .gitignore file to prevent committing it:

.idea/
.venv/
.env

Load environmental variables by adding the following code:

from dotenv import load_dotenv

load_dotenv()

• L3: Load environment variables from the .env file

Security Warning: Never hardcode your API key in any of your Python files nor commit the .env file to GitHub.

Test API Connection

Install the openai package:

pip install openai

Verify if your API key works correctly:

Code

import os

from dotenv import load_dotenv
from openai import OpenAI
from openai.types.chat import ChatCompletion

load_dotenv()

def test_llm_api() -> ChatCompletion:
    client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))

    response = client.chat.completions.create(
        model="gpt-5-nano",
        messages=[
            {"role": "user", "content": "Say 'Hello World'."}
        ]
    )

    return response

if __name__ == "__main__":
    r = test_llm_api()
    print(r.choices[0].message.content)
    print(r)

• L1: Import the os package

• L5: ChatCompletion class

• L9: Use typing to indicate the return type (ChatCompletion).

• L10: Initialize the OpenAI client.

• L12-17: Create a simple chat completion

  • L13: Use the fastest and cheapest model for testing

Output

Here are common issues and troubleshooting you may encounter:

Issue	Solution
AuthenticationError	Your API key is invalid or not set. Double-check the key and environment variable name.
RateLimitError	You are making requests too quickly. Add delays between requests or upgrade your account tier.
InsufficientQuotaError	You have exceeded your usage limits or need to add billing information. Check your usage page.
ModuleNotFoundError: openai	The OpenAI package isn't installed. Run pip install openai in the Terminal.
Environment variable not found	Make sure you have set up the environment variable correctly and restarted PyCharm if needed.

References

1. Source: src/llm_api_access.py