Update app.py

app.py

@@ -35,10 +35,6 @@ model = AutoModelForCausalLM.from_pretrained(
     torch_dtype=torch.bfloat16,
 ).to(device)
 
-if device == "cuda":
-    model = torch.compile(model)
-
-
 @spaces.GPU
 def stream_chat(
     message: str,