Exploring SDKs through Python

Calling Google Gemini with google.genai

• The code imports genai from google and creates a genai.Client() instance.
• load_dotenv() is called before creating the client to load configuration from environment variables.
• client.models.generate_content() is used with model="gemini-2.5-flash" and the movie prompt as contents.
• The response text from Gemini is printed to the console using print(response.text).

Code Implementation:

from dotenv import load_dotenv
from google import genai

load_dotenv()

client = genai.Client()

response = client.models.generate_content(
    model="gemini-2.5-flash",
    contents="Suggest one good movie for software engineers"
)

print(response.text)

Calling Local Models with Ollama Chat API

• The code imports chat and ChatResponse from the ollama package.
• The chat() function is called with model='mistral' and a messages list containing the user movie prompt.
• The returned value is stored in a ChatResponse object referenced by the variable response.
• The model output is printed twice using response['message']['content'] and response.message.content.

Code Implementation:

from ollama import chat
from ollama import ChatResponse

response: ChatResponse = chat(
    model='mistral',
    messages=[
        {
            'role': 'user',
            'content': 'Suggest one good movie for software engineers',
        },
    ]
)

print(response['message']['content'])
print(response.message.content)

Calling OpenAI with the Python SDK

• The code imports OpenAI from openai and calls load_dotenv() before creating client = OpenAI().
• client.responses.create() is used with model="gpt-5-nano" and the same movie prompt passed in the input field.
• The API call returns a response object that provides the generated text through response.output_text.
• The generated movie name is displayed by printing response.output_text to the console.

Code Implementation:

from dotenv import load_dotenv
from openai import OpenAI

load_dotenv()

client = OpenAI()

response = client.responses.create(
    model="gpt-5-nano",
    input="Suggest one good movie for software engineers"
)

print(response.output_text)

Common Pattern Across All

• Each example first creates a client object specific to the provider before making any request.
• All three code samples send the same simple English prompt asking for one suitable movie for software engineers.
• The result from the model is stored in a variable named response.
• The final step in each example is to print the response so the model output is visible in the terminal.