import gradio as gr
import websockets
import asyncio
import json
import base64
from PIL import Image
import io

def process_image_stream(image, question, max_tokens):
    return "This is a test response"

# Create Gradio interface
demo = gr.Interface(
    fn=process_image_stream,
    inputs=[
        gr.Image(type="filepath", label="Upload Image"),
        gr.Textbox(
            label="Question", 
            placeholder="Ask a question about the image...",
            value="Describe this image"
        ),
        gr.Slider(
            minimum=50,
            maximum=200,
            value=200,
            step=1,
            label="Max Tokens"
        )
    ],
    outputs=gr.Textbox(label="Response", interactive=False),
    title="Nexa Omni Vision",
    description=f"""
    Model Repo: <a href="https://huggingface.co/NexaAIDev/omnivision-968M">NexaAIDev/omnivision-968M</a>

    *Model updated on Nov 21, 2024\n
    Upload an image and ask questions about it. The model will analyze the image and provide detailed answers to your queries.
    """,
)

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