sbai 0.1.2

Simplified AI API wrapper for ChatGPT and Google Gemini, enabling easy integration of language models with configurable parameters

sbai (Simple Broader AI)

A Python package to simplify interactions with OpenAI and Google Gemini APIs.

Installation

pip install sbai

Features

• Easy-to-use interfaces for ChatGPT and Gemini
• Configurable model parameters
• Optional conversation memory
• Support for system instructions

Usage

ChatGPT Integration

from sbai import ChatGPT

# Initialize with OpenAI API key
client = ChatGPT(
    api_key="your_openai_api_key", 
    model="gpt-4o-mini", 
    memory=True
)

# Send a prompt
response = client.prompt("Write a haiku about AI")
print(response)

# Clear conversation history
client.clear_history()

Gemini Integration

from sbai import Gemini

# Initialize with Google API key
client = Gemini(
    api_key="your_google_api_key", 
    model="gemini-2.0-flash-exp", 
    instructions="Be concise and helpful"
)

# Send a prompt
response = client.prompt("Explain quantum computing")
print(response)

# Clear conversation history
client.clear_history()

Configuration Options

ChatGPT

• api_key: Your OpenAI API key (required)
• model: OpenAI model to use (default: "gpt-4o-mini")
• temperature: Sampling temperature (0-1, default: 0)
• top_p: Nucleus sampling parameter (0-1, default: 0)
• instructions: System-level instructions
• max_token: Maximum tokens in response (default: 200)
• memory: Enable conversation memory (default: False)
• memory_limit: Maximum messages in memory (default: 5)

Gemini

• api_key: Your Google API key (required)
• model: Gemini model to use (default: "gemini-2.0-flash-exp")
• temperature: Sampling temperature (0-1, default: 0)
• top_p: Nucleus sampling parameter (0-1, default: 0)
• max_token: Maximum tokens in response (default: 200)
• memory: Enable conversation memory (default: False)
• instructions: System-level instructions

Requirements

• Python 3.8+
• openai library
• google-generativeai library

License

MIT License