app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline

model_name = "anasmkh/customized_llama3.1_8b"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype=torch.float16)

generator = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    max_new_tokens=64,
    temperature=1.5,
    min_p=0.1,
)

def chat(message, history):
    history = history or []
    history.append({"role": "user", "content": message})
    response = generator(history)[-1]["generated_text"]
    history.append({"role": "assistant", "content": response})
    return history

with gr.Blocks() as demo:
    chatbot = gr.Chatbot()
    message = gr.Textbox()
    clear = gr.ClearButton([message, chatbot])

    message.submit(chat, [message, chatbot], chatbot)
    clear.click(lambda: None, None, chatbot, queue=False)

demo.launch()