Update app.py

app.py

@@ -111,7 +111,10 @@ if prompt := st.chat_input(f"Hi I'm {selected_model}, How can I help you today?"
         "ModernMigrate⭐": 8192,
         "RetroRecode🔄": 4096
     }
-    max_new_tokens = max_tokens[selected_model] - len(formated_text.split())
+    
+    # Calculate available tokens for new content
+    input_tokens = len(formated_text.split())
+    max_new_tokens = max(0, max_tokens[selected_model] - input_tokens)
     
     with st.chat_message("assistant"):
         client = InferenceClient(