# import gradio as gr
# from huggingface_hub import InferenceClient

# """
# For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
# """
# client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")


# def respond(
#     message,
#     history: list[tuple[str, str]],
#     system_message,
#     max_tokens,
#     temperature,
#     top_p,
# ):
#     messages = [{"role": "system", "content": system_message}]

#     for val in history:
#         if val[0]:
#             messages.append({"role": "user", "content": val[0]})
#         if val[1]:
#             messages.append({"role": "assistant", "content": val[1]})

#     messages.append({"role": "user", "content": message})

#     response = ""

#     for message in client.chat_completion(
#         messages,
#         max_tokens=max_tokens,
#         stream=True,
#         temperature=temperature,
#         top_p=top_p,
#     ):
#         token = message.choices[0].delta.content

#         response += token
#         yield response


# """
# For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
# """
# demo = gr.ChatInterface(
#     respond,
#     additional_inputs=[
#         gr.Textbox(value="You are a friendly Chatbot.", 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()

import gradio as gr
from huggingface_hub import InferenceClient
from PIL import Image
import io
import base64

client = InferenceClient("meta-llama/Llama-3.2-11B-Vision-Instruct")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
    image: Image,  # Add image input to the function
):
    # Prepare the system message and history for the conversation
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    # Add the current user message
    messages.append({"role": "user", "content": message})

    # Convert the image to a base64-encoded string
    image_bytes = io.BytesIO()
    image.save(image_bytes, format='PNG')
    image_bytes.seek(0)
    image_base64 = base64.b64encode(image_bytes.getvalue()).decode('utf-8')

    # Use InferenceClient to handle the image and text input to the model
    # Pass the base64-encoded image as the input
    response_data = client.text_to_image(images=image_base64, prompt=message)  # Pass the base64 string as 'images'

    # Check if the response is in the correct format (e.g., image)
    try:
        # Assuming the response is an image in base64 format
        if 'image' in response_data:
            image_response = response_data['image']
            print("Image Res:")
            print(image_response)
            # Decode the base64 image back into an image object
            image_bytes = base64.b64decode(image_response)
            image = Image.open(io.BytesIO(image_bytes))
            image.show()  # Or return the image in Gradio
            return "Image processed successfully"  # You can return some confirmation or process the image further
        else:
            return "Error: No valid image returned from the model."
    except Exception as e:
        return f"Error processing image: {e}"

# Create the Gradio interface with an image input
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", 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)"),
        gr.Image(type="pil", label="Upload an Image"),  # Image input for vision tasks
    ],
)

if __name__ == "__main__":
    demo.launch(share=True)  # Set share=True to create a public link