app.py

import gradio as gr
import random
from huggingface_hub import InferenceClient


client = InferenceClient("HuggingFaceH4/zephyr-7b-beta") #change the LLM
def respond(message, history):
    messages = [{"role": "system", "content": "You are a professional interviewer."}]
    if history:
        messages.extend(history)
    messages.append({"role":"user", "content": message})
    response = ""
    for message in client.chat_completion(
        messages,
        max_tokens = 100, #change the length
        temperature = 0.5,
        top_p = 0.7,
        stream=True
    ):
        token = message.choices[0].delta.content
        response += token
        yield response

def echo(message, history):
    choices = ["Without a doubt", "Ask again later","Yes, definitely","Don't count on it"]
    response = random.choice(choices)
    return response
  
with gr.Blocks(css="""
#title { font-size: 28px; color: #00e676; text-align: center; }
.message.user { background-color: #1e88e5; color: white; border-radius: 12px; }
.message.bot { background-color: #424242; color: #eeeeee; border-radius: 12px; }
.chatbot { background-color: #121212; padding: 10px; border-radius: 16px; }
body { background-color: #111 !important; }
""") as demo:
    
    gr.Markdown("# <div id='title'>🤖 AI Study Buddy</div>")
    
    gr.ChatInterface(
        respond,
        type="messages",
        description="Ask me anything about AI, ML, or Python!",
        submit_btn="Send",
        stop_btn="Stop",
        chatbot=gr.Chatbot(show_label=False, elem_classes=["chatbot"])
    )

demo.launch(debug=True)
chatbot.launch(debug=True)