Update app.py

app.py

@@ -37,11 +37,12 @@ this demo is governed by the original [license](https://github.com/meta-llama/ll
 #     DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
 
 
-if torch.cuda.is_available():
+if torch.cuda.is_available() or os.getenv("ZERO_GPU_SUPPORT", False):
     model_id = "chuanli11/Llama-3.2-3B-Instruct-uncensored"
     model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype=torch.bfloat16)
     tokenizer = AutoTokenizer.from_pretrained(model_id)
-    # tokenizer.use_default_system_prompt = False
+else:
+    raise RuntimeError("No compatible GPU environment found for this model.")
 
 
 @spaces.GPU