import gradio as gr
import torch
from PIL import Image
from transformers import AutoModel, AutoTokenizer

# Load the model and tokenizer
model = AutoModel.from_pretrained('openbmb/MiniCPM-Llama3-V-2_5', trust_remote_code=True, torch_dtype=torch.float16)
model = model.to(device='cuda')
tokenizer = AutoTokenizer.from_pretrained('openbmb/MiniCPM-Llama3-V-2_5', trust_remote_code=True)
model.eval()

# Define a function to generate a response
def generate_response(image, question):
    msgs = [{'role': 'user', 'content': question}]
    res = model.chat(
        image=image,
        msgs=msgs,
        tokenizer=tokenizer,
        sampling=True,
        temperature=0.7,
        stream=True
    )
    generated_text = ""
    for new_text in res:
        generated_text += new_text
    return generated_text


# Create a Gradio interface
iface = gr.Interface(
    fn=generate_response,
    inputs=[gr.Image(type="pil"), "text"],
    outputs="text",
    title="Visual Question Answering",
    description="Input an image and a question related to the image to receive a response.",
)

# Launch the app
iface.launch(debug=True)