Lesson 4: AI Chatbot with Groq API

What is Groq?

Groq is an AI inference platform that provides extremely fast responses using optimized hardware (LPU - Language Processing Unit).

Key Benefits:

• Very fast inference speeds (responses in milliseconds)
• Free tier available for developers
• Access to models like Llama, Mixtral, and more
• Simple API similar to OpenAI

Getting Started with Groq

Step 1: Sign up for a Groq account at console.groq.com

Step 2: Generate an API key from your dashboard

Step 3: Install the Groq Python library:

pip install groq

Chatbot Blueprint Structure

Your chatbot blueprint should have these routes:

from flask import Blueprint, render_template, request

chatbot_bp = Blueprint('chatbot', __name__)

@chatbot_bp.route('/chatbot')
def index():
    # Display chatbot interface
    return render_template('chatbot/index.html')

@chatbot_bp.route('/chatbot/ask', methods=['POST'])
def ask_question():
    question = request.form.get('question')
    response = get_groq_response(question)
    return render_template('chatbot/index.html', 
                         question=question, 
                         response=response)

Integrating Groq API

Here's how to use Groq to generate AI responses:

from groq import Groq
import os

def get_groq_response(question):
    # Initialize Groq client
    client = Groq(
        api_key=os.environ.get('GROQ_API_KEY')
    )
    
    # Create chat completion
    chat_completion = client.chat.completions.create(
        messages=[
            {
                "role": "user",
                "content": question,
            }
        ],
        model="llama3-8b-8192",  # or mixtral-8x7b-32768
    )
    
    return chat_completion.choices[0].message.content

Available Models:

• llama3-8b-8192 - Fast, good for general questions
• llama3-70b-8192 - More powerful, better reasoning
• mixtral-8x7b-32768 - Excellent balance of speed and quality

Chatbot Form Implementation

Create a simple form for users to ask questions:

<form action="/chatbot/ask" method="POST">
    <div class="mb-3">
        <label for="question" class="form-label">Ask a Question</label>
        <textarea class="form-control" name="question" 
                  id="question" rows="3" 
                  placeholder="e.g., When is Christmas this year?"
                  required></textarea>
    </div>
    <button type="submit" class="btn btn-primary">
        Send Question
    </button>
</form>

{% if response %}
<div class="alert alert-success mt-4">
    <h5>Response:</h5>
    <p>{{ response }}</p>
</div>
{% endif %}

Example Conversation

User asks: "When is Christmas this year?"

Groq responds: "Christmas is on December 25th. So it's on a Monday."

Tips for better responses:

• Be specific with your questions
• Use llama3-70b for complex questions
• Handle API errors gracefully (show user-friendly messages)
• Consider adding a system message for consistent personality

Environment Variables

Store your Groq API key securely using environment variables:

Create a .env file:

GROQ_API_KEY=gsk_your_api_key_here

Load it in your Flask app:

from dotenv import load_dotenv
import os

load_dotenv()

# Access the key
api_key = os.environ.get('GROQ_API_KEY')

Important: Never commit your .env file to GitHub! Add it to .gitignore.

Module 6 Complete!

Congratulations! You've now learned how to build all four blueprints:

• Tickers: Stock price tracking with financial API
• Weather: City weather tracking with weather API
• Movies: Movie database with OMDB API and full CRUD
• Chatbot: AI-powered Q&A with Groq API

Next, you'll learn theory concepts and then see the complete demo of all four working together!