app.py

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

# Load model and tokenizer with CPU-compatible settings
model_name = "davnas/Italian_Cousine_2.1"
tokenizer = AutoTokenizer.from_pretrained(model_name)

# Configure quantization properly
quantization_config = BitsAndBytesConfig(
    load_in_4bit=False,
    load_in_8bit=False,
    bnb_4bit_quant_type=None
)

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    device_map="cpu",  # Explicitly set to CPU
    torch_dtype=torch.float32,
    quantization_config=quantization_config,
    use_safetensors=True,
    low_cpu_mem_usage=True,
)

def respond(message, history, system_message, max_tokens, temperature, top_p):
    # Format the conversation
    messages = [{"role": "system", "content": system_message}]
    
    # Add history
    for user_msg, assistant_msg in history:
        messages.append({"role": "user", "content": user_msg})
        messages.append({"role": "assistant", "content": assistant_msg})
    
    # Add current message
    messages.append({"role": "user", "content": message})
    
    # Create the prompt using the tokenizer's chat template
    input_ids = tokenizer.apply_chat_template(
        messages,
        tokenize=True,
        add_generation_prompt=True,
        return_tensors="pt"
    )
    
    # Generate response
    with torch.no_grad():
        output_ids = model.generate(
            input_ids,
            max_new_tokens=max_tokens,
            do_sample=True,
            temperature=temperature,
            top_p=top_p,
            pad_token_id=tokenizer.pad_token_id,
        )
    
    # Decode and return the response
    response = tokenizer.decode(output_ids[0][len(input_ids[0]):], skip_special_tokens=True)
    return response

# Create the interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(
            value="You are a professional chef assistant who provides accurate and detailed recipes.",
            label="System message"
        ),
        gr.Slider(
            minimum=1,
            maximum=2048,
            value=512,
            step=1,
            label="Max new tokens"
        ),
        gr.Slider(
            minimum=0.1,
            maximum=4.0,
            value=0.7,
            step=0.1,
            label="Temperature"
        ),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)"
        ),
    ],
    title="Italian Cuisine Chatbot",
    description="Ask me anything about Italian cuisine or cooking!"
)

if __name__ == "__main__":
    demo.launch(server_name="0.0.0.0", server_port=7860)