app.py

import gradio as gr
import random
import time

# Use a pipeline as a high-level helper
# Use a pipeline as a high-level helper
from transformers import pipeline

print("loading model")
pipe = pipeline("conversational", model="llSourcell/medllama2_7b")
print("load model")
with gr.Blocks() as demo:
    chatbot = gr.Chatbot()
    msg = gr.Textbox()
    clear = gr.ClearButton([msg, chatbot])

    def respond(message, chat_history):
        bot_message = pipe(message)
        chat_history.append((message, bot_message))
        time.sleep(2)
        return "", chat_history

    msg.submit(respond, [msg, chatbot], [msg, chatbot])

demo.launch()