Update app.py

app.py

@@ -47,7 +47,7 @@ def chat(message, history, temperature, do_sample, max_tokens):
             chat.append({"role": "assistant", "content": item[1]})
     chat.append({"role": "user", "content": message})
     messages = tok.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
-    model_inputs = tok([messages], return_tensors="pt").to(device)
+    model_inputs = tok([messages], return_tensors="pt",return_token_type_ids=False).to(device)
     streamer = TextIteratorStreamer(
         tok, timeout=20.0, skip_prompt=True, skip_special_tokens=True
     )