🤖 gemini_helper

Use Gemini API with OpenAI Agent SDK easily! This package allows students and developers to use free Gemini models (like gemini-2.0-flash) with the powerful OpenAI Agent SDK.

🚀 Installation

Using uv (recommended):

uv add gemini-agent-helper

Or via pip:

pip install gemini-agent-helper

⚙️ Environment Setup

Create a .env file in your project root and add your Gemini API key:

GEMINI_API_KEY=your_gemini_api_key_here

📌 You can get your key from Google AI Studio

💡 How to Use

✅ Default (uses gemini-2.0-flash)

from gemini_helper.core import get_gemini_model

model = get_gemini_model()

⚡ Custom model (like gemini-1.0-pro)

model = get_gemini_model("gemini-1.0-pro")

💻 Full Example

from gemini_helper.core import get_gemini_model
from agents import Agent, Runner

model = get_gemini_model()

agent = Agent(
    name="Helper",
    instructions="You are a helpful assistant",
    model=model
)

result = Runner.run_sync(agent, input="What is Python?")
print(result.final_output, end="")  

📆 Features

• 🔐 Secure API key handling via .env
• 🔄 Compatible with OpenAI Agent SDK
• 💸 No need for OpenAI's paid API – fully free using Gemini
• ⚙️ Optional model switching (e.g. gemini-1.0-pro, gemini-2.0-pro)
• 🧹 Easy to integrate into agent-based workflows

🛠️ Dependencies

This package depends on:

• openai-agent-sdk
• python-dotenv

These are automatically installed with the package.

🧑‍💻 Author

Made with ❤️ by Huriya Syed Empowering students to build free agents using Gemini + OpenAI SDK.

📬 Contact / Support

Feel free to open an issue or connect with me on LinkedIn Or email me at: huriyasyed462@gmail.com