Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# 🌟 Gemini 2.5 Pro — API Proxy Model
|
| 2 |
+
|
| 3 |
+
This repository hosts a **Custom Hugging Face Inference Endpoint** that connects directly to **Google Gemini** through the official `google-generativeai` SDK.
|
| 4 |
+
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
## 🧠 Overview
|
| 8 |
+
|
| 9 |
+
This model is **not an open-weight model**, but a proxy wrapper around the official Google Gemini API.
|
| 10 |
+
|
| 11 |
+
It allows developers to deploy a **Gemini-powered chatbot** on Hugging Face with:
|
| 12 |
+
- Full `handler.py` control
|
| 13 |
+
- Secure API key injection via Environment Variables
|
| 14 |
+
- Compatible with any frontend (React, Flask, HTML, etc.)
|
| 15 |
+
- Optional CORS for browser-based chatbots
|
| 16 |
+
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
## ⚙️ How It Works
|
| 20 |
+
|
| 21 |
+
When deployed, the endpoint:
|
| 22 |
+
1. Receives a POST request (`/`) with either:
|
| 23 |
+
```json
|
| 24 |
+
{"inputs": "Hello Gemini!"}
|