update

app.py

@@ -109,6 +109,19 @@ with gr.Blocks() as demo:
     gr.Markdown("## VTBench")
     gr.Markdown("---")
 
+    gr.Markdown("👋 **Welcome to VTBench!** Upload an image, select models, and click 'Start Processing' to compare results side by side.")
+    with gr.Accordion("📘 Full Instructions", open=False):
+        gr.Markdown("""
+**VTBench User Guide**
+
+- **Upload an image** or click one of the example images.
+- **Select one or more models** from the list.
+- Click **Start Processing** to run inference.
+- Selected model outputs appear first, others show placeholders.
+
+⚠️ *Each model is downloaded on first use. Please wait patiently the first time you run a model.*
+""")
+
     image_input = gr.Image(
         type="pil",
         label="Upload an image",
@@ -138,6 +151,7 @@ with gr.Blocks() as demo:
 
     gr.Markdown("---")
     gr.Markdown("⚠️ **The more models you select, the longer the processing time will be.**")
+    gr.Markdown("*Note: Each model is downloaded on first use. Subsequent uses will load from cache and run faster.*")
 
     display_names = list(model_name_mapping.values())
     default_selected = ["SD3.5L", "Chameleon", "Janus Pro 1B/7B"]