How to Build a Flask ChatGPT API (with Free Hosting Options)

Build a Flask ChatGPT API: Step-by-Step Guide with Free Hosting

If you’re looking to create a simple yet powerful backend service to interact with ChatGPT using Flask, you’re in the right place. In this guide, I’ll show you how to build a lightweight Flask ChatGPT API, and we’ll go through how to host it online—for free.

Step 1: Project Setup

Start by creating a new folder for your project and a virtual environment:

mkdir flask-chat-api
cd flask-chat-api
python -m venv venv
source venv/bin/activate  # On Windows use venv\Scripts\activate

Install the required packages:

pip install flask openai python-dotenv

Create the following file structure:

flask-chat-api/
│
├── app.py
├── .env
└── requirements.txt

Add your dependencies to requirements.txt:

flask
openai
python-dotenv

Step 2: Configure Environment Variables

In your .env file, store your OpenAI API key securely:

OPENAI_API_KEY=your_openai_key_here

We’ll use this key inside the app to authenticate requests to OpenAI.

Step 3: Create the Flask App

Now open app.py and add the following code:

from flask import Flask, request, jsonify
from dotenv import load_dotenv
import openai
import os
 
load_dotenv()
 
app = Flask(__name__)
openai.api_key = os.getenv("OPENAI_API_KEY")
 
@app.route('/chat', methods=['POST'])
def chat():
    data = request.get_json()
    user_message = data.get("message")
 
    if not user_message:
        return jsonify({"error": "No message provided"}), 400
 
    try:
        response = client.chat.completions.create(
            model="gpt-3.5-turbo",
            messages=[
                {"role": "system", "content": "You are a helpful assistant."},
                {"role": "user", "content": user_message}
            ]
        )
        reply = response.choices[0].message.content
        return jsonify({"response": reply})
    except Exception as e:
        return jsonify({"error": str(e)}), 500
 
if __name__ == '__main__':
    app.run(debug=True)

Step 4: Run Your API Locally

To test your API, run:

flask run

Your API will be available at http://127.0.0.1:5000/chat. You can test it with a tool like Postman or curl:

curl -X POST http://127.0.0.1:5000/chat \
     -H "Content-Type: application/json" \
     -d '{"message": "Hello, who are you?"}'

Hosting Your Flask API

Once your Flask API is ready, you might want to deploy it online so others (or other services) can access it. Here are some great hosting platforms that support Flask and offer generous free tiers:

1. PythonAnywhere

A simple platform to host Python apps.
Easy setup with a browser-based environment.
Great for small-scale APIs and testing.
Free plan available.

2. Koyeb

Modern serverless platform.
Supports Docker and Git-based deployments.
Fast global infrastructure.
Free tier with generous limits.

3. Render

Very easy to use, supports Git deployments.
Great documentation and developer-friendly setup.
Free plan with limited resources.

4. Google App Engine

Part of Google Cloud Platform.
Offers a free tier that works well for low-traffic apps.
Supports Python natively.

5. Fly.io

Deploy applications close to your users.
Good performance and flexibility.
Free tier includes small instances, perfect for testing APIs.

These platforms are great for deploying small Flask APIs like the one we just built. Choose the one that best fits your needs based on simplicity, performance, and scalability.

What’s Next?

You now have a fully working Flask ChatGPT API that you can run locally or host online. If you want to expand this, consider:

Adding logging or analytics
Rate-limiting or API keys
Using GPT-4 instead of GPT-3.5
Securing your endpoints with authentication