Update app.py

app.py

@@ -72,7 +72,7 @@ def respond(
     response = ""
 
     for message in tokenizer.batch_decode(model.generate(
-        input_ids=input_ids,
+        input_ids=input_ids_list,
         max_new_tokens=max_tokens,
         do_sample=True,
         top_p=top_p,