plugllm 0.1.1

Unified LLM API interface for OpenAI, Gemini, Mistral, Groq etc.

🔌 plugllm

plugllm is a unified and provider-agnostic Python package that lets you interact with multiple LLM APIs (like OpenAI, Gemini, Mistral, Groq, etc.) using a single, consistent interface — without needing to learn each provider’s SDK.

Created by Yash Kumar Firoziya

---

🌟 Features

• 🔌 Unified API — One interface for all providers
• 📡 Supports multiple providers — OpenAI, Gemini, Mistral, Groq (more coming)
• 🧠 Same request structure — Compatible message format across providers
• 🔐 Secure & simple config — Use environment variables or inline setup
• 🚫 No SDKs required — Only uses Python requests library
• 📜 Role-based prompt support — For both single and multi-turn chat
• 🔄 Extensible — Add custom providers easily

---

📦 Installation

pip install plugllm

---

⚙️ Configuration

You can configure directly in your code:

from plugllm import config

config(
    provider="openai",       # "gemini", "mistral", "groq" also supported
    api_key="your-api-key",
    model="gpt-4",           # model name based on provider
    base_url=None            # optional: custom or local API endpoint
)

Or use environment variables for better security:

export LLM_PROVIDER=openai
export LLM_API_KEY=your-api-key
export LLM_MODEL=gpt-4

---

💬 Basic Usage

from plugllm import generate

response = generate("What is quantum entanglement?")
print(response)

🧵 Multi-turn Chat

generate([
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What are black holes?"}
])

---

📡 Supported Providers

• OpenAI (ChatGPT, GPT-4, GPT-3.5)
• Google Gemini
• Mistral AI
• GroqCloud (Mixtral)
• Coming soon: Cohere, Anthropic Claude, Ollama, LM Studio

---

🗂️ Project Structure

plugllm/
├── __init__.py
├── core.py
├── config.py
├── prompts.py
└── providers/
    ├── base.py
    ├── openai.py
    ├── gemini.py
    ├── mistral.py
    └── groq.py

---

🤝 Contributing

Pull requests are welcome! If you want to add support for a new provider, just create a new module in providers/ based on base.py.

---

🪪 License

This project is licensed under the MIT License.

---

✨ Author

Yash Kumar Firoziya

---