| library_name: fastapi | |
| tags: | |
| - qwen | |
| - openai-api | |
| - jwt | |
| - inference | |
| license: mit | |
| # Qwen FastAPI Inference API | |
| This repository contains a secure, OpenAI-compatible REST API for the **Qwen2.5-0.5B-Instruct** model, built with FastAPI. | |
| ## Features | |
| - **OpenAI Compatible**: Implements the `/v1/chat/completions` endpoint. | |
| - **Secure**: Uses JWT Authentication (HS256) for all inference requests. | |
| - **Lightweight**: Optimized to run on consumer hardware or free-tier cloud environments like Google Colab. | |
| ## Structure | |
| - `application.py`: The main FastAPI server. | |
| - `client.py`: A test client to verify the API. | |
| - `generate_token.py`: Utility to generate JWT tokens for authentication. | |