Openai

pip install datapizza-ai-clients-openai

datapizza.clients.openai.OpenAIClient

Bases: Client

A client for interacting with the OpenAI API.

This class provides methods for invoking the OpenAI API to generate responses based on given input data. It extends the Client class.

__init__

__init__(
    api_key,
    model="gpt-4o-mini",
    system_prompt="",
    temperature=None,
    cache=None,
    base_url=None,
    organization=None,
    project=None,
    webhook_secret=None,
    websocket_base_url=None,
    timeout=None,
    max_retries=2,
    default_headers=None,
    default_query=None,
    http_client=None,
)

Parameters:

Name	Type	Description	Default
api_key	str	The API key for the OpenAI API.	required
model	str	The model to use for the OpenAI API.	'gpt-4o-mini'
system_prompt	str	The system prompt to use for the OpenAI API.	''
temperature	float | None	The temperature to use for the OpenAI API.	None
cache	Cache | None	The cache to use for the OpenAI API.	None
base_url	str | URL | None	The base URL for the OpenAI API.	None
organization	str | None	The organization ID for the OpenAI API.	None
project	str | None	The project ID for the OpenAI API.	None
webhook_secret	str | None	The webhook secret for the OpenAI API.	None
websocket_base_url	str | URL | None	The websocket base URL for the OpenAI API.	None
timeout	float | Timeout | None	The timeout for the OpenAI API.	None
max_retries	int	The max retries for the OpenAI API.	2
default_headers	dict[str, str] | None	The default headers for the OpenAI API.	None
default_query	dict[str, object] | None	The default query for the OpenAI API.	None
http_client	Client | None	The http_client for the OpenAI API.	None

Usage example

from datapizza.clients.openai import OpenAIClient

client = OpenAIClient(
    api_key="YOUR_API_KEY",
    model="gpt-4o-mini",
)
response = client.invoke("Hello!")
print(response.text)

Include thinking

from datapizza.clients.openai import OpenAIClient

client = OpenAIClient(
    model= "gpt-5",
    api_key="YOUR_API_KEY",
)

response = client.invoke("Hi",reasoning={
        "effort": "low",
        "summary": "auto"
    }
)
print(response)