LLama 3.1 based chatbot example.

python/community/llama-bot/README.md

@@ -0,0 +1,43 @@
+# Llama 3.1 Chatbot with Flet 
+A simple chatbot application using Llama 3.1 model from Groq, built with the Flet framework. This app provides a user interface for sending messages to the chatbot and receiving responses.
+## Installation
+
+1. Clone the repository:
+
+    ```bash
+    git clone https://github.com/yourusername/llama-chatbot.git
+    cd llama-chatbot
+    ```
+
+2. Install the required Python packages:
+
+    ```bash
+    pip install flet groq
+    ```
+
+3. Set up your Groq API key. Follow the instructions below to obtain your API key.
+
+## Getting the API Key from Groq
+
+1. Visit the [Groq website](https://groq.com) and sign up for an account.
+2. Navigate to the API section of your account settings.
+3. Generate a new API key and copy it.
+
+## Usage
+
+1. Open the `chatbot.py` file in a text editor.
+2. Replace the placeholder API key with your Groq API key:
+
+    ```python
+    client = Groq(
+        api_key='your_groq_api_key_here',
+    )
+    ```
+
+3. Save the file and run the application:
+
+    ```bash
+    python chatbot.py
+    ```
+
+4. The application will open a window with the chat interface. Type your message and press "Send" to interact with the chatbot.

python/community/llama-bot/app.py

@@ -0,0 +1,59 @@
+from groq import Groq
+import flet as ft
+
+# Initialize Groq client with API key
+client = Groq(
+    api_key='your_groq_api_key_here',
+)
+
+class Message:
+    def __init__(self, user: str, text: str, response_text: str):
+        self.user = user
+        self.text = text
+        self.response_text = response_text
+
+def main(page: ft.Page):
+    chat = ft.ListView(expand=True, spacing=10, padding=10, auto_scroll=True)
+    new_message = ft.TextField(expand=True, hint_text="Type your message here...")
+
+    def on_message(message: Message):
+        chat.controls.append(ft.Text(f"User: {message.text}"))
+        chat.controls.append(ft.Text(f"Bot: {message.response_text}"))
+        page.update()
+
+    page.pubsub.subscribe(on_message)
+
+    def send_click(e):
+        user_message = new_message.value
+        if user_message:
+            new_message.value = ""
+            processing_text = ft.Text("Processing answer...", color="blue")
+            chat.controls.append(processing_text)
+            page.update()
+            chat_completion = client.chat.completions.create(
+                messages=[
+                    {
+                        "role": "user",
+                        "content": user_message,
+                    }
+                ],
+                model="llama-3.1-70b-versatile",
+            )
+            response_text = chat_completion.choices[0].message.content
+            message = Message(user=page.session_id, text=user_message, response_text=response_text)
+            page.pubsub.send_all(message)
+            new_message.value = ""
+            page.update()
+
+    page.add(
+        ft.Container(
+            content=ft.Column([
+                chat,
+                ft.Row([new_message, ft.ElevatedButton("Send", on_click=send_click)])
+            ]),
+            expand=True,
+            padding=10
+        )
+    )
+
+ft.app(target=main)

python/community/llama-bot/requirements.txt

@@ -0,0 +1,2 @@
+flet
+groq