Update app.py

app.py

@@ -38,7 +38,15 @@ def load_model():
 def clean_text(text):
     model, tokenizer = load_model()
     
-    inputs = tokenizer(text, return_tensors="pt", max_length=4096, truncation=True)
+    # Apply chat template
+    messages = [
+        {"role": "user", "content": text}
+    ]
+    
+    # Apply the chat template
+    formatted_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+    
+    inputs = tokenizer(formatted_text, return_tensors="pt", max_length=4096, truncation=True)
     inputs = {k: v.cuda() for k, v in inputs.items()}
     
     # Enable streaming