A Developer’s Info To OpenAI’s GPT-5 Model Capabilities

!pip arrange pandas openai

import os
from getpass import getpass
os.environ['OPENAI_API_KEY'] = getpass('Enter OpenAI API Key: ')

from openai import OpenAI
import pandas as pd
from IPython.present import present

shopper = OpenAI()

question = "Write a poem a number of detective and his first treatment"

info = []

for verbosity in ["low", "medium", "high"]:
    response = shopper.responses.create(
        model="gpt-5-mini",
        enter=question,
        textual content material={"verbosity": verbosity}
    )

    # Extract textual content material
    output_text = ""
    for merchandise in response.output:
        if hasattr(merchandise, "content material materials"):
            for content material materials in merchandise.content material materials:
                if hasattr(content material materials, "textual content material"):
                    output_text += content material materials.textual content material

    utilization = response.utilization
    info.append({
        "Verbosity": verbosity,
        "Sample Output": output_text,
        "Output Tokens": utilization.output_tokens
    })

# Create DataFrame
df = pd.DataFrame(info)

# Present correctly with centered headers
pd.set_option('present.max_colwidth', None)
styled_df = df.kind.set_table_styles(
    [
        {'selector': 'th', 'props': [('text-align', 'center')]},  # Coronary heart column headers
        {'selector': 'td', 'props': [('text-align', 'left')]}     # Left-align desk cells
    ]
)

present(styled_df)

from openai import OpenAI

shopper = OpenAI()

response = shopper.responses.create(
    model="gpt-5-mini",
    enter="Please use the code_exec instrument to calculate the cube of the number of vowels inside the phrase 'pineapple'",
    textual content material={"format": {"type": "textual content material"}},
    devices=[
        {
            "type": "custom",
            "name": "code_exec",
            "description": "Executes arbitrary python code",
        }
    ]
)

print(response.output[1].enter)

from openai import OpenAI
import re

shopper = OpenAI()

email_regex = r"^[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+.[A-Za-z]{2,}$"

quick = "Give me a sound e mail deal with for John Doe. It might be a dummy e mail"

# No grammar constraints -- model might give prose or invalid format
response = shopper.responses.create(
    model="gpt-4o",  # or earlier
    enter=quick
)

output = response.output_text.strip()
print("GPT Output:", output)
print("Official?", bool(re.match(email_regex, output)))

from openai import OpenAI

shopper = OpenAI()

email_regex = r"^[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+.[A-Za-z]{2,}$"

quick = "Give me a sound e mail deal with for John Doe. It might be a dummy e mail"

response = shopper.responses.create(
    model="gpt-5",  # grammar-constrained model
    enter=quick,
    textual content material={"format": {"type": "textual content material"}},
    devices=[
        {
            "type": "custom",
            "name": "email_grammar",
            "description": "Outputs a valid email address.",
            "format": {
                "type": "grammar",
                "syntax": "regex",
                "definition": email_regex
            }
        }
    ],
    parallel_tool_calls=False
)

print("GPT-5 Output:", response.output[1].enter)

import time
from openai import OpenAI

shopper = OpenAI()

quick = "Classify the given amount as odd and even. Return one phrase solely."

start_time = time.time()  # Start timer

response = shopper.responses.create(
    model="gpt-5",
    enter=[
        { "role": "developer", "content": prompt },
        { "role": "user", "content": "57" }
    ],
    reasoning={
        "effort": "minimal"  # Faster time-to-first-token
    },
)

latency = time.time() - start_time  # End timer

# Extract model's textual content material output
output_text = ""
for merchandise in response.output:
    if hasattr(merchandise, "content material materials"):
        for content material materials in merchandise.content material materials:
            if hasattr(content material materials, "textual content material"):
                output_text += content material materials.textual content material

print("--------------------------------")
print("Output:", output_text)
print(f"Latency: {latency:.3f} seconds")