FastAPI LLM Backend

A simple FastAPI backend for LLM-based features.

The project has a clean API route, request validation, environment configuration, and an LLM service layer. If no API key is available, it returns a demo response so the backend can still be tested locally.

Features

• FastAPI /generate endpoint
• request and response models
• optional OpenAI API support
• demo mode without an API key
• simple health check route
• Dockerfile included

Tech Stack

• Python
• FastAPI
• Pydantic
• OpenAI API optional
• Docker

Setup

python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt

Create .env if using OpenAI:

OPENAI_API_KEY=your_api_key_here

Run

uvicorn app.main:app --reload

Open:

http://127.0.0.1:8000/docs

Example Request

{
  "prompt": "Explain embeddings in simple words",
  "temperature": 0.3
}