🚀 GemmaLangServeAPI

GemmaLangServeAPI is a FastAPI-powered API service built using Google's Gemma 2B LLM integrated through LangChain and LangServe. It allows users to submit Python code and receive human-like explanations in response — ideal for educational, debugging, or AI-assistance use cases.

Hosted via ngrok for easy sharing and accessible testing through Postman.

---

📌 Features

• 🧠 Explains Python functions in natural language
• 🔗 Easily deployable via FastAPI + LangServe
• ⚡ Supports Hugging Face Gemma 2B model pipeline
• 🧪 API testing-ready via Postman or Swagger UI
• ☁️ Hosted via Google Colab with public URL through ngrok

---

🛠️ Tech Stack

Layer	Technology / Tool
API Framework	FastAPI
LLM	Gemma 2B by Google
Model Pipeline	HuggingFace Transformers
Orchestration	LangChain + LangServe
Serving	Uvicorn
Tunneling	ngrok
LLM Wrapper	HuggingFacePipeline
Test Tool	Postman
Platform	Google Colab

---

🚀 How to Run

🧱 Prerequisites

Install dependencies:

pip install fastapi uvicorn transformers langchain langserve langchain_huggingface huggingface_hub pyngrok nest_asyncio

▶️ Run the Server

python app.py

You’ll get a public URL via ngrok, like:

🔗 https://your-subdomain.ngrok-free.app/docs

Open the /docs route to explore Swagger UI or test it using Postman.

---

📮 Postman API Guide

Method	Endpoint	Description
POST	/explain	Submit Python code for explanation

🧾 Headers

Content-Type: application/json

📤 Request Body

{
  "code": "def add(a, b): return a + b"
}

✅ Example Response

{
  "output": "This function takes two parameters and returns their sum."
}

---

🧩 Project Structure

GemmaLangServeAPI/
├── app.py            # FastAPI + LangChain setup
├── requirements.txt  # All dependencies
└── README.md         # You're reading it!

---

📚 Use Cases

• 👨‍💻 Educational tools or tutorials
• 🧠 AI-powered code explanation service
• 🤖 Backend service for developer assistance tools

---

📝 License

This project is licensed under the MIT License.

---

🙏 Acknowledgements

• 🤖 Google AI – Gemma LLM
• 🧱 LangChain
• 🤗 Hugging Face
• ⚡ FastAPI