Update app.py

app.py

@@ -12,7 +12,7 @@ tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_ID,
     load_in_8bit=True,  # or try load_in_4bit=True
-    device_map="cpu"
+    device_map=DEVICE
 )
 
 #########