app.py

import gradio as gr
import random
from huggingface_hub import InferenceClient

client = InferenceClient("HuggingFaceH4/zephyr-7b-beta", provider='hf-inference')

def respond(message, history):

    system_message = "You are a sweet chatbot that gives good and caring advice."

    
    messages = [{"role": "system", "content": system_message}]

    if history:
        messages.extend(history)

    messages.append({"role": "user", "content": message})

    response = ""
    
    for message in client.chat_completion(
        messages,
        max_tokens=300,
        temperature=1.3,
        top_p=0.4,
        stream=True
    ):
        token = message.choices[0].delta.content
        response += token
        yield response

chatbot = gr.ChatInterface(respond, type='messages', title="My First Chatbot!")
chatbot.launch()