Lesson 3 · OpenAI

Mastering OpenAI Models: Enhance AI Responses with Python Embeddings

Welcome to your guide for getting started You will learn step by step how to use the OpenAI API Let us get started! The OpenAI API lets you interact with…

⬇ Download notebook Open in Colab ↗

CourseOpenAI
Lesson3 of 14
Video20 min
FormatJupyter notebook · 14 code cells

NumPy OpenAI

Data

No separate download needed — the notebook creates or downloads everything it uses.

📓 Full notebook

Download .ipynb

Learn the OpenAI API in Python#

Welcome to your guide for getting started
You will learn step by step how to use the OpenAI API
Let us get started!

What is the OpenAI API?#

The OpenAI API lets you interact with advanced AIs using Python
You can generate text, summarize, answer questions, work with images, and more
OpenAI's models help you build apps that can "think" like a human

import warnings; warnings.filterwarnings("ignore")
import numpy as np
np.random.seed(42)
# Suppress warning messages for a cleaner experience

Setting up: Install OpenAI library#

Before you can use OpenAI, make sure the library is installed
Usually, you run pip install openai in the terminal
We will use it directly in this notebook

# Import the required OpenAI library
from openai import OpenAI

# Create an OpenAI client instance
client = OpenAI()

About your API key#

Your API key lets you use OpenAI services
It is stored securely in environment settings
You do NOT need to type in a key during this lesson

# Access your OpenAI API key from environment settings
import os
API_key = os.environ["OPENAI_API_KEY"]

What can the OpenAI API do?#

Generate chat messages and answers
Summarize and rewrite text
Turn speech into text and vice versa
Analyze or embed information into numbers
Power chatbots, assistants, and smart apps

Core idea: Chat completions#

The chat completion is how you talk to the model
You send messages, and the AI sends back a reply
You choose the model size: large models are smarter but slower

# Basic Chat Completion
response = client.chat.completions.create(
    model='gpt-4.1-mini',
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

Hello! How can I help you today?

Try it: Ask anything!#

Let us ask the AI something new in your own words
You can type any question or prompt you want
The AI will do its best to answer

# Prompt the user for a question to ask the model
user_prompt = input("Type your question to AI: ")
response2 = client.chat.completions.create(
    model='gpt-4.1-mini',
    messages=[{"role": "user", "content": user_prompt}]
)
print(response2.choices[0].message.content)

AI is increasingly integrated into many aspects of daily life, enhancing convenience, efficiency, and personalization. Here are some common uses of AI in everyday activities:

1. **Virtual Assistants:** AI-powered assistants like Siri, Google Assistant, and Alexa help with tasks such as setting reminders, checking the weather, playing music, and answering questions.

2. **Personalized Recommendations:** Streaming services (Netflix, Spotify), shopping platforms (Amazon), and social media (Instagram, Facebook) use AI to suggest content, products, or connections based on your preferences and behavior.

3. **Smart Home Devices:** AI enables smart thermostats, lights, security cameras, and appliances to learn your habits and optimize energy use, enhance security, or improve convenience.

4. **Navigation and Travel:** GPS apps (Google Maps, Waze) use AI to analyze traffic data in real-time and provide efficient routes or estimated arrival times.

5. **Communication:** Email services use AI to filter spam and categorize messages; predictive text and speech recognition improve typing and accessibility.

6. **Health and Fitness:** Wearable devices and apps track activity, monitor vital signs, and provide personalized health insights or coaching.

7. **Financial Management:** AI helps detect fraudulent transactions, manage budgets, and provide investment advice via robo-advisors.

8. **Customer Service:** AI chatbots provide instant support on websites and apps, answering common questions or troubleshooting issues.

9. **Photography:** Smartphone cameras use AI for scene recognition, enhancing image quality, and facial recognition for organizing photos.

10. **Education:** AI-powered apps offer personalized learning experiences, language translation, and tutoring support.

These examples show how AI makes daily tasks easier, more efficient, and more tailored to individual needs.

Quick tip: Choosing models#

OpenAI offers different models for speed and power
Small models are fast and cheap
Large models can answer complex questions but may cost more

# Try changing the model to a bigger one
response_big = client.chat.completions.create(
    model='gpt-4.1',
    messages=[{"role": "user", "content": "Tell me a fun fact about space."}]
)
print(response_big.choices[0].message.content)

Sure! Here’s a fun fact: 

**A day on Venus is longer than a year on Venus!**  
It takes Venus about 243 Earth days to rotate once on its axis (that's one Venus day), but only about 225 Earth days to orbit the Sun (one Venus year). So, surprisingly, a day on Venus is longer than its year!

# Handle model errors gracefully
try:
    bad_response = client.chat.completions.create(
        model='nonexistent-model',
        messages=[{"role": "user", "content": "test"}]
    )
except Exception as error:
    print("There was an error: ", error)

There was an error:  Error code: 404 - {'error': {'message': 'The model `nonexistent-model` does not exist or you do not have access to it.', 'type': 'invalid_request_error', 'param': None, 'code': 'model_not_found'}}

Other OpenAI API abilities#

Summarization: shrink long text to short answers
Embeddings: turn sentences into vector numbers
Function calling: ask AI to trigger your code functions
Speech: convert voice to text, or text to audio

# Summarize a passage using the API
long_text = """OpenAI has developed language models that are helpful for students, teachers, and anyone who works with information. These tools can explain things, write emails, and help you find answers fast."""
summary = client.chat.completions.create(
    model='gpt-4.1-mini',
    messages=[{"role": "user", "content": f"Summarize: {long_text}"}]
)
print(summary.choices[0].message.content)

OpenAI has created language models that assist students, teachers, and information workers by explaining concepts, writing emails, and quickly providing answers.

# Get an embedding for a short sentence
embed = client.embeddings.create(
    model='text-embedding-3-small',
    input=['Machine learning is fun']
)
print(len(embed.data[0].embedding))

# Use function calling to let AI add numbers
def add(a, b):
    return a + b

resp = client.chat.completions.create(
    model='gpt-4.1',
    messages=[{"role": "user", "content": "Add 3 and 5"}],
    functions=[
        {
            "name": "add",
            "description": "Add two numbers",
            "parameters": {
                "type": "object",
                "properties": {
                    "a": {"type": "number"},
                    "b": {"type": "number"}
                },
                "required": ["a", "b"]
            }
        }
    ],
    function_call='auto'
)
print(resp)

ChatCompletion(id='chatcmpl-Chf5GI8K6dVRSsFgf1uihbkbWp3Mx', choices=[Choice(finish_reason='function_call', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=[], audio=None, function_call=FunctionCall(arguments='{"a":3,"b":5}', name='add'), tool_calls=None))], created=1764522066, model='gpt-4.1-2025-04-14', object='chat.completion', service_tier='default', system_fingerprint='fp_09249d7c7b', usage=CompletionUsage(completion_tokens=17, prompt_tokens=50, total_tokens=67, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=0)))

# Convert speech to text (requires a .wav file named sample.wav)
audio_file = open('sample.wav', 'rb')
transcript = client.audio.transcriptions.create(
    model='whisper-1',
    file=audio_file
)
print(transcript.text)

Welcome to OpenAI.

# Convert text to speech (requires playing audio)
with open('voice.mp3', 'wb') as f:
    audio = client.audio.speech.create(
        model='gpt-4o-mini-tts',
        voice='alloy',
        input='Hello world!'
    )
    f.write(audio.read())

print("Audio saved as voice.mp3")

Audio saved as voice.mp3

PROJECT: Your own chatbot#

Let us build a simple chatbot that replies to your questions
This will show how to use a chat loop with OpenAI
You can keep chatting until you say stop

# Simple chatbot loop: ask until you type 'stop'
print("Chatbot online! Type 'stop' to end.")
while True:
    user_msg = input("You: ")
    if user_msg.strip().lower() == 'stop':
        print("Chatbot stopped.")
        break
    reply = client.chat.completions.create(
        model='gpt-4.1-mini',
        messages=[{"role": "user", "content": user_msg}]
    )
    print("AI:", reply.choices[0].message.content)

Chatbot online! Type 'stop' to end.
AI: I'm doing great, thank you! How can I assist you today?
AI: Sure! Here's one for you:

Why did the scarecrow win an award?  
Because he was outstanding in his field!
Chatbot stopped.

Troubleshooting tips#

If you see errors, check your API key or spelling
If responses are slow, try a smaller model
If audio does not work, check the file name and format

Challenge: Generate a vacation plan#

Write code to ask the AI for a 3-day vacation itinerary to Paris
Print the result

Challenge: Help with study tips#

Use the API to generate 5 helpful study tips
Try customizing the output style

Recap: What you have learned#

How to set up and use the OpenAI API
How chat completions work
How to handle errors and troubleshoot
How to use advanced features like speech and functions

Thanks for learning OpenAI in Python!#

Try more ideas and keep exploring
Subscribe for more coding lessons
Happy coding!

Found this useful?

All lessons, notebooks and datasets here are free. If they helped you, a coffee keeps new lessons coming.

☕ Buy me a coffee Other ways to help