app.py

import gradio as gr
import random
from huggingface_hub import InferenceClient

client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

def respond(message, history):

    messages = [{"role": "system", "content": "You are a friendly chatbot."}]
    if history:
        messages.extend(history)
        
    messages.append({"role": "user", "content": message})
    response = client.chat_completion(
        messages,
        max_tokens=100
    )
    print(response['choices'][0]['message']['content'].strip())
    return response['choices'][0]['message']['content'].strip()
    
def echo(message, history):
    choices = ["yes", "no", "better not tell you now", "without a doubt", "very doubtful", "cannot predict now", "you may rely on it", "don't count on it", "reply hazy, try again"]
    new = random.choice(choices)
    return new
    
chatbot = gr.ChatInterface(respond, type = "messages", title = "Chatbot")

chatbot.launch()