Update app.py

app.py

@@ -301,12 +301,21 @@ demo = gr.ChatInterface(
         gr.Slider(label="Repetition penalty", minimum=1.0, maximum=2.0, step=0.05, value=1.2),
     ],
     examples=[
-        [{"text": "Create a short story based on the images.", "files": ["examples/1111.jpg", "examples/2222.jpg", "examples/3333.jpg"]}],
+        [
+            {
+                "text": "Create a short story based on the images.",
+                "files": [
+                    "examples/1111.jpg",
+                    "examples/2222.jpg",
+                    "examples/3333.jpg",
+                ],
+            }
+        ],
         [{"text": "Explain the Image", "files": ["examples/3.jpg"]}],
         [{"text": "Explain the content of the Advertisement", "files": ["examples/videoplayback.mp4"]}],
         [{"text": "Which movie character is this?", "files": ["examples/9999.jpg"]}],
         ["Explain Critical Temperature of Substance"],
-        [{"text": "Transcription of the letter", "files": ["examples/222.png"]}],
+        [{"text": "@qwen2-vl Transcription of the letter", "files": ["examples/222.png"]}],
         [{"text": "Explain the content of the video in detail", "files": ["examples/breakfast.mp4"]}],
         [{"text": "Describe the video", "files": ["examples/Missing.mp4"]}],
         [{"text": "Explain what is happening in this video ?", "files": ["examples/oreo.mp4"]}],
@@ -317,7 +326,7 @@ demo = gr.ChatInterface(
     ],
     cache_examples=False,
     type="messages",
-    description="# **Gemma 3 Multimodal**",
+    description="# **Gemma 3 Multimodal \n`Use @qwen2-vl to switch to Qwen2-VL OCR for image inference and `@video-infer` for video input.`**",
     fill_height=True,
     textbox=gr.MultimodalTextbox(label="Query Input", file_types=["image", "video"], file_count="multiple", placeholder="Tag with @qwen2-vl for Qwen2-VL inference if needed."),
     stop_btn="Stop Generation",