app.py

import gradio as gr
import os
import io
from PIL import Image
from google import genai
from google.genai import types


# ==============================
# 1. Gemini Client
# ==============================
client = genai.Client(
    api_key=os.environ.get("GEMINI_API_KEY")
)


# ==============================
# 2. Multimodal Function (SAFE)
# ==============================
def analyze_image(image, prompt):

    try:
        # ---- Checks ----
        if image is None:
            return "❌ No image uploaded"

        if not prompt or prompt.strip() == "":
            return "❌ Prompt is empty"

        # Convert image to bytes
        buffer = io.BytesIO()
        image.save(buffer, format="PNG")
        image_bytes = buffer.getvalue()

        image_part = types.Part.from_bytes(
            data=image_bytes,
            mime_type="image/png"
        )

        # ✅ USE FLASH MODEL (IMPORTANT)
        response = client.models.generate_content(
            model="gemini-1.5-flash",
            contents=[
                prompt,
                image_part
            ]
        )

        return response.text

    except Exception as e:
        # 🔍 Show real Gemini error
        return f"❌ Gemini Error:\n{str(e)}"


# ==============================
# 3. Gradio UI
# ==============================
interface = gr.Interface(
    fn=analyze_image,
    inputs=[
        gr.Image(type="pil", label="Upload Image"),
        gr.Textbox(label="Prompt", value="Describe the image")
    ],
    outputs=gr.Textbox(label="Response"),
    title="Gemini Multimodal Test App",
)


# ==============================
# 4. Launch
# ==============================
interface.launch(ssr_mode=False)