Update app.py

app.py

@@ -19,7 +19,7 @@ if not torch.cuda.is_available():
     raise ValueError("Running on CPU 🥶 This demo does not work on CPU.")
 
 model_id = "neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50"
-model = LLM(model_id, max_model_len=MAX_INPUT_TOKEN_LENGTH)
+model = LLM(model_id, sparsity="sparse_w16a16", max_model_len=MAX_INPUT_TOKEN_LENGTH)
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 tokenizer.use_default_system_prompt = False