app.py

import gradio as gr
from huggingface_hub import InferenceClient

# Initialize the client
client = InferenceClient(
    model="davnas/Italian_Cousine_2.1",
    headers={"Content-Type": "application/json"}
)

def respond(message, history, system_message, max_tokens, temperature, top_p):
    # Format the prompt including history and system message
    prompt = ""
    
    # Add system message if provided
    if system_message:
        prompt += f"{system_message}\n"
    
    # Add conversation history
    for msg in history:
        if isinstance(msg, list) and len(msg) == 2:
            prompt += f"User: {msg[0]}\nAssistant: {msg[1]}\n"
    
    # Add current message
    prompt += f"User: {message}\nAssistant:"
    
    # Prepare parameters for text generation
    parameters = {
        "max_new_tokens": max_tokens,
        "temperature": temperature,
        "top_p": top_p,
        "return_full_text": False
    }
    
    response = ""
    try:
        # Use generate_text with proper parameters
        for token in client.text_generation(
            prompt,
            stream=True,
            **parameters
        ):
            response += token
            yield response
    except Exception as e:
        yield f"Error: {str(e)}"

# Create the interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(
            value="You are a helpful assistant knowledgeable about Italian cuisine.",
            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)"
        ),
    ]
)

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