Update app.py

app.py

@@ -1,66 +1,75 @@
-
 import os
 from google import genai
 from google.genai import types
 from PIL import Image
 from io import BytesIO
 import gradio as gr
-import PIL.Image
 
+# Set up your API key
 API_KEY = os.getenv("GOOGLE_API_KEY")
 client = genai.Client(api_key=API_KEY)
 
 def edit_image_with_gemini(image, text_input):
     """
     Edits an image using Gemini 2.0 Flash Experimental API based on a given text prompt.
-    Parameters:
-        image_path (str): Path to the input image.
-        text_prompt (str): Text prompt describing the edit.
-    Returns:
-        Image: The modified image.
     """
-
     if image is None or not text_input:
         return "Please upload an image and provide an edit prompt.", None
 
-    # Generate content with Gemini API
-    response = client.models.generate_content(
-        model="gemini-2.0-flash-exp-image-generation",
-        contents=[text_input, image],
-        config=types.GenerateContentConfig(
-            response_modalities=['Text', 'Image']
+    # Wrap the text prompt as a tuple (as shown in the docs)
+    text_prompt_tuple = (text_input,)
+    
+    try:
+        response = client.models.generate_content(
+            model="gemini-2.0-flash-exp-image-generation",
+            contents=[text_prompt_tuple, image],
+            config=types.GenerateContentConfig(
+                response_modalities=['Text', 'Image']
+            )
         )
-    )
+    except Exception as e:
+        return f"API call error: {e}", None
 
-    # Extract and display the generated image
+    # Process the response: check for both image and text output
     for part in response.candidates[0].content.parts:
         if part.inline_data is not None:
             edited_image = Image.open(BytesIO(part.inline_data.data))
             return "Here is your edited image:", edited_image
+        elif part.text is not None:
+            # Optionally, print or log the text output
+            print("Text output:", part.text)
 
     return "No image was generated. Try modifying your prompt.", None
 
+def generate_thumbnail_prompt():
+    """
+    Returns a predefined optimized prompt for creating a YouTube thumbnail.
+    """
+    return ("Generate a bold, eye-catching YouTube thumbnail with vibrant colors, "
+            "large text, and a strong contrast. Make sure it stands out and is attention-grabbing.")
 
-
-# Gradio App
+# Build the Gradio interface
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
     gr.Markdown("# ✨ AI-Powered Image Editor with Gemini 2.0 Flash Experimental")
     gr.Markdown("Upload an image and describe the edits you want!")
-
+    
     with gr.Row():
         image_input = gr.Image(type="pil", label="Upload Image")
         text_input = gr.Textbox(placeholder="Describe your edit...", label="Edit Prompt")
-
+    
+    thumbnail_master_btn = gr.Button("🎨 Thumbnail Master")
     output_text = gr.Textbox(label="Status", interactive=False)
     output_image = gr.Image(label="Edited Image")
 
     with gr.Row():
         submit_btn = gr.Button("Generate Edit")
         clear_btn = gr.Button("Clear")
+
     text_input.submit(edit_image_with_gemini, [image_input, text_input], [output_text, output_image])
     submit_btn.click(edit_image_with_gemini, [image_input, text_input], [output_text, output_image])
     clear_btn.click(lambda: (None, None), None, [output_text, output_image])
+    # Use inputs=None since generate_thumbnail_prompt requires no input
+    thumbnail_master_btn.click(generate_thumbnail_prompt, inputs=None, outputs=text_input)
 
-# Launch the app
 if __name__ == "__main__":
     demo.launch(debug=True)