Update app_chat.py

app_chat.py

@@ -26,7 +26,7 @@ DESCRIPTION = """\
 model_id = "nvidia/Hymba-1.5B-Instruct"
 model = AutoModelForCausalLM.from_pretrained(model_id, device_map="cuda", trust_remote_code=True)
 model = model.cuda().to(torch.bfloat16)
-
+model.compile()
 #model.to('cuda')
 #model.eval()
 tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)