Update app.py

app.py

@@ -53,23 +53,43 @@ def predict_image(img):
 
 # 5. Build and launch the Web App!
 description_text = """
-### How it works:
+
+###  How it works:
+
 This model analyzes microscopic pixel noise to determine if an image is real or AI-generated. 
 
+
+
 ###  Limitations for best results:
-* **Resolution Sweet Spot:** Works best on smaller images (around **256x256 to 512x512 pixels**). Ultra-high resolution (1080p or 4K) images will fail because the model's 'magnifying glass' is only 224x224 pixels, meaning it gets lost looking at a tiny fraction of the photo!
-* **Centered Subjects:** The model strictly scans the dead-center of the image. If the AI mistakes are on the edges, it won't see them!
-* **No Screenshots:** Heavy compression (like taking a screenshot or downloading from social media) destroys the forensic evidence. Please upload raw files.
+
+* **Resolution Sweet Spot:** Works flawlessly on standard AI resolutions and mid-sized images (from **512x512 up to around 1000x1500 pixels**, like 640x832 or 880x1320). 
+
+* **The 4K Danger Zone:** Ultra-high-resolution (like **3840x2160 / 4K**) images will cause the model to fail. Because the model's 'magnifying glass' is strictly fixed to a 224x224 pixel crop, it ends up looking through a pinhole at less than 0.6% of a 4K image, causing it to lose context and guess randomly.
+
+* **Centered Subjects:** The model strictly scans the dead-center of the image. If the AI artifacts or mistakes (like extra fingers or warped backgrounds) are on the far edges, the model won't see them!
+
+* **No Screenshots:** Heavy compression (like taking a screenshot or downloading from messaging apps) destroys the microscopic forensic evidence. Please upload the raw, original files.
+
 """
 
+
+
 # Inside your interface:
+
 interface = gr.Interface(
+
     fn=predict_image,
+
     inputs=gr.Image(type="pil", label="Upload an Image"),
+
     outputs=gr.Label(num_top_classes=2, label="Prediction"),
+
     title="PixelSleuth: AI Image Detector",
+
     description=description_text, 
+
     flagging_mode="never"
+
 )
 
 interface.launch()