Update app.py

app.py

@@ -60,4 +60,19 @@ print("OCR OUTPUT: ", ocr_inference(img, "korean"))
 words=text_inference(text, "korean")
 print("TEXT INPUT: ", text)
 print("WORD PARSING: ",words)
-print("flashcard output:",make_flashcards(words, "korean"))
+print("flashcard output:",make_flashcards(words, "korean"))
+
+
+demo = gr.ChatInterface(
+    fn=model_inference,
+    description="# **Multimodal OCR `@RolmOCR and Default Qwen2VL OCR`**",
+    examples=examples,
+    textbox=gr.MultimodalTextbox(
+        label="Query Input", 
+        file_types=["image", "video"], 
+        file_count="multiple", 
+        placeholder="Use tag @RolmOCR for RolmOCR, or leave blank for default Qwen2VL OCR"
+    ),
+    stop_btn="Stop Generation",
+    multimodal=True,
+    cache_examples=False,