app.py

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

# Load tokenizer and model
model_id = "HuggingFaceH4/zephyr-7b-beta"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

# Define the Gradio interface
with gr.Blocks(fill_height=True) as demo:
    with gr.Sidebar():
        gr.Markdown("## Zephyr-7B Unlimited Assistant")
        gr.Markdown(
            "This assistant is powered by the HuggingFaceH4/zephyr-7b-beta model.\n"
            "You can start chatting right away!"
        )
        login_button = gr.LoginButton("🔐 Sign in to Hugging Face")  # Optional UI

    chatbot = gr.Chatbot(label="🧠 Zephyr-7B Assistant")
    user_input = gr.Textbox(placeholder="Ask anything...", show_label=False)

    chat_history = []

    def chat(user_msg, history):
        # Add system + user messages to chat history
        messages = [
            {"role": "system", "content": "You are a friendly chatbot who always responds in the style of a pirate."}
        ]
        for human, ai in history:
            messages.append({"role": "user", "content": human})
            messages.append({"role": "assistant", "content": ai})
        messages.append({"role": "user", "content": user_msg})

        # Format the prompt using the tokenizer's chat template
        prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)

        # Generate response
        outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
        response = outputs[0]["generated_text"].split("</s>")[-1].strip()

        # Append new interaction
        history.append((user_msg, response))
        return history, ""

    user_input.submit(chat, inputs=[user_input, chatbot], outputs=[chatbot, user_input])

demo.launch()