Creating a Simple Chatbot using OpenAI API using Python

.env.example

# Rename this file to .env and replace the value below with your actual API key
OPENAI_API_KEY="your-api-key-here"

README.md

Simple Chatbot using OpenAI API with Python

This is a basic example of how to build a command-line chatbot using the OpenAI API and Python.

Requirements

This project requires:

• Python 3.9.1 or earlier
• An OpenAI API Key (How to get your API key)

Setup

1. Clone or Create Your Project Folder

Create a new folder where you want to run this chatbot example.

2. Add the Required Files

Create or copy the following files into your project folder:

• chatbot.py – Main Python script
• requirements.txt – Lists required Python packages
• .env – Stores your API key (you can start with the .env.example provided)

3. Install Dependencies

Install the required packages using pip:

$ pip install -r requirements.txt

4. Add Your OpenAI API Key

Create a .env file (or rename .env.example), and add your API key:

OPENAI_API_KEY="your-api-key-here"

How to Run

To start the chatbot, run the following command:

$ python chatbot.py

Then just start chatting! Type exit to quit.

---

Example

Welcome to Chatbot!
You: Hello!
Assistant: Hi there! How can I assist you today?

---

License

This project is open source and available under the MIT License.

Let me know if you want to add sections like screenshots, architecture diagram, or deployment instructions later!

chatbot.py

from typing import Dict, List

import openai
from .setttings import settings

openai_client = openai.Client(api_key=settings.OPENAI_API_KEY)


def get_gpt_response(
        current_messages: List[Dict],
        model: str = settings.OPENAI_CHATGPT_MODEL,
        max_tokens: int = settings.OPENAI_CHATGPT_MAX_TOKENS,
        temperature: int = settings.OPENAI_CHATGPT_TEMPERATURE,
        stream: bool = settings.OPENAI_CHATGPT_STREAM_RESPONSE,
) -> List:
    response = openai_client.chat.completions.create(
        messages=current_messages,
        model=model,
        max_tokens=max_tokens,
        temperature=temperature,
        stream=stream
    )
    if stream:
        full_response = ''
        print('Assistant: ', end='')
        for chunk in response:
            text = chunk.choices[0].delta.content
            if text:
                full_response += text
                print(text, end='')
        print()
    else:
        full_response = response.choices[0].message.content
        print(full_response)
    current_messages.append({'role': 'assistant', 'content': full_response})
    return current_messages

if __name__ == '__main__':
    print(
        'Welcome to Chatbot! \n'
        'Assistant: Hi there! How can I assist you today?')
    messages = []
    while True:
        question = input('User: ')
        if not question or question in ['quit', 'exit']:
            print('Assistant: See you later!')
            break

        messages.append({'role': 'user', 'content': question})
        messages = get_gpt_response(messages)

requirements.txt

openai
prettyconf

settings.py

from prettyconf import config

class Settings:
    OPENAI_API_KEY = config('OPENAI_API_KEY', default='')
    OPENAI_CHATGPT_MODEL = config('OPENAI_CHATGPT_MODEL', default='gpt-3.5-turbo-0125')
    OPENAI_CHATGPT_MAX_TOKENS = config('OPENAI_CHATGPT_MAX_TOKENS', default=1000, cast=int)
    OPENAI_CHATGPT_TEMPERATURE = config('OPENAI_CHATGPT_TEMPERATURE', default=0, cast=int)
    OPENAI_CHATGPT_STREAM_RESPONSE = config('OPENAI_CHATGPT_STREAM_RESPONSE', default=True, cast=config.boolean)


settings = Settings()