app.py

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

# Load the model and tokenizer from the local path
model = AutoModel.from_pretrained('openbmb/MiniCPM-Llama3-V-2_5', trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained('openbmb/MiniCPM-Llama3-V-2_5', trust_remote_code=True)

# Set the model to evaluation mode
model.eval()

def predict(image, question):
    # Preprocess the image
    image = image.convert('RGB')

    # Create the message list
    msgs = [{'role': 'user', 'content': question}]

    # Generate a response
    res = model.chat(
        image=image,
        msgs=msgs,
        tokenizer=tokenizer,
        sampling=True,
        temperature=0.1
    )
    return res

# Create the Gradio interface
iface = gr.Interface(
    fn=predict,
    inputs=[
        gr.inputs.Image(type="pil", label="Upload an Image"),
        gr.inputs.Textbox(label="Ask a Question")
    ],
    outputs="text",
    title="Image Question Answering",
    description="Upload an image and ask a question about it."
)

# Launch the app
iface.launch()