Update app.py

app.py

@@ -21,14 +21,14 @@ logger = logging.getLogger(__name__)
 # Initialize BitNet model and tokenizer
 try:
     model_name = "1bitLLM/bitnet_b1_58-3B"
-    # Fallback to LLaMA tokenizer
     tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/llama-tokenizer", cache_dir="/app/cache")
     model = AutoModelForCausalLM.from_pretrained(
         model_name,
         torch_dtype=torch.float32,
         device_map="cpu",
+        low_cpu_mem_usage=True,
         cache_dir="/app/cache",
-        trust_remote_code=True  # Allow custom model code
+        trust_remote_code=True
     )
     pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=512)
 except Exception as e: