Basic interaction with OpenAI GPT models

This example demonstrates how to set up an OpenAI GPT model and send a basic user prompt with JavaScript.

Requirements

• Node.js
• An OpenAI API key

Steps for running this example in the shell

1. Install dependencies:

npm install

2. Export your OpenAI API key as an environment variable:

export OPENAI_API_KEY="sk-..." # Windows cmd: set OPENAI_API_KEY="sk-..." # Windows PowerShell: $env:OPENAI_API_KEY="sk-..."

3. Run the script:

npm start

Output

When you run the script, it will send a user prompt to a GPT model (gpt-4o-mini) using the temperature parameter. Then, it will send the same user prompt to a more advanced model (gpt-5) using reasoning. The output will show the responses from both models.

=== Basic model  ===
User: How many tokens are in your context window?
        Model: gpt-4o-mini-2024-07-18
        Latency: 1.695 seconds
        Input tokens: 15
        Output tokens: 39
        Reasoning tokens: 0
        Total tokens: 54
AI: My context window can handle up to 8,192 tokens. This includes both the input and the output tokens. If you have any specific questions or need assistance, feel free to ask!

=== Advanced model  ===
User: How many tokens are in your context window?
        Model: gpt-5-2025-08-07
        Latency: 17.243 seconds
        Input tokens: 14
        Output tokens: 750
        Reasoning tokens: 640
        Total tokens: 764
AI: It depends on the model. Most current ChatGPT models available here use a 128,000‑token context window (shared across your input and my output). If you tell me the specific model you’re using, I can confirm the exact limit.