Aydie's Avenue Logo Aydie's Avenue

aydie-genai 1.1.0

$ pip install aydie-genai

Project description

aydie_genai is a Python library designed to eliminate the complexity of working with multiple Generative AI models. Instead of writing different code for Gemini, OpenAI, Claude, Groq, and others, you can use one simple, unified function to access them all.

Key Features

  • Unified Interface: A single generate() function for all supported models.
  • Simple & Intuitive: Get started in minutes. No need to learn multiple SDKs.
  • Provider Agnostic: Switch between models like gpt-4o and gemini-1.5-pro by changing a single string.
  • Built-in Documentation: Rich docstrings provide in-console help via help(genai.generate).
  • Robust Error Handling: Custom exceptions for common issues like missing API keys.

Supported Models

Provider Supported Models Environment Variable
Google gemini-1.5-pro, gemini-1.5-flash, gemini-1.0-pro... GOOGLE_API_KEY
OpenAI gpt-4o, gpt-4-turbo, gpt-3.5-turbo... OPENAI_API_KEY
Anthropic claude-3-opus-20240229, claude-3-sonnet-20240229... ANTHROPIC_API_KEY
Groq llama3-70b-8192, llama3-8b-8192, mixtral-8x7b-32768... GROQ_API_KEY
DeepSeek deepseek-chat, deepseek-coder, deepseek-v2-chat DEEPSEEK_API_KEY

⚡ Quick Start

1. Set Up Your API Keys

The library loads API keys from environment variables. The easiest way to manage these is to create a .env file in your project's root directory.

# .env file
GOOGLE_API_KEY="your_google_api_key"
OPENAI_API_KEY="your_openai_api_key"
ANTHROPIC_API_KEY="your_anthropic_api_key"
GROQ_API_KEY="your_groq_api_key"
DEEPSEEK_API_KEY="your_deepseek_api_key"

2. Generate a Response

Using the library is incredibly simple. Just import genai and call the generate function.

from aydie_genai import genai

# Example 1: Using Google Gemini
response_gemini = genai.generate(
    model='gemini-1.5-pro',
    prompt='Explain the theory of relativity in simple terms for a five-year-old.',
    system_instruction='You are a friendly and patient teacher.'
)
print("Gemini says:", response_gemini)

# Example 2: Switching to OpenAI's GPT-4o is as easy as changing the model name
response_openai = genai.generate(
    model='gpt-4o',
    prompt='Write a short story about a robot who discovers music.'
)
print("\nOpenAI says:", response_openai)
							

Function Parameters

The genai.generate() function accepts the following parameters:

Parameter Type Description
model str Required. The identifier for the model you want to use (e.g., 'gpt-4o').
prompt str Required. The main user input or question for the model.
system_instruction str A directive to guide the model's behavior or personality.
temperature float Controls randomness (0.0 for deterministic, 2.0 for random). Defaults to 1.0.
max_tokens int The maximum number of tokens to generate in the response. Defaults to 2048.
top_p float Nucleus sampling parameter. Defaults to 1.0.
api_key str Directly pass an API key, overriding the environment variable. Defaults to None.

Project links

Meta

  • License: MIT License
  • Author: Aydie