app.py

import os
import gradio as gr
import google.generativeai as genai

# Initialize the Gemini API
def initialize_gemini():
    api_key = os.getenv("GOOGLE_API_KEY")
    if not api_key:
        raise ValueError("GOOGLE_API_KEY is not set in environment variables.")
    genai.configure(api_key=api_key)
    return genai.GenerativeModel("gemini-2.0-flash-exp")  # Assuming gemini-2.0-flash-exp is the correct model name

# Handle file inputs and text to generate a response
def generate_response(text_input, files):
    contents = []
    if text_input:
        contents.append({"text": text_input})  # Add text input to the contents
    
    for file in files:
        with open(file.name, "rb") as f:
            data = f.read()
        # Guess the MIME type from the file input
        mime_type = gr.io.utils.guess_mime_type(file.name)
        if mime_type:
            contents.append({
                "mime_type": mime_type,
                "data": data
            })

    model = initialize_gemini()
    response = model.generate_content(contents)
    response.resolve()
    return response.text

# Gradio Interface
def setup_interface():
    with gr.Blocks() as app:
        with gr.Row():
            text_input = gr.Textbox(label="Enter any text:", placeholder="Type your question or comment here...")
            file_input = gr.File(label="Upload your files", file_types=["image", "video", "audio", "pdf"])
            submit_button = gr.Button("Generate Response")
        response_output = gr.Textbox(label="Response from Gemini", placeholder="Response will appear here...")

        submit_button.click(
            fn=generate_response,
            inputs=[text_input, file_input],
            outputs=[response_output]
        )

    return app

# Run the interface
if __name__ == "__main__":
    app = setup_interface()
    app.launch(share=True)  # Set `share=True` for Hugging Face Spaces or public deployment