diff --git a/python/llm/example/NPU/HF-Transformers-AutoModels/LLM/minicpm.py b/python/llm/example/NPU/HF-Transformers-AutoModels/LLM/minicpm.py
index a0d92a38..79786167 100644
--- a/python/llm/example/NPU/HF-Transformers-AutoModels/LLM/minicpm.py
+++ b/python/llm/example/NPU/HF-Transformers-AutoModels/LLM/minicpm.py
@@ -37,6 +37,12 @@ if __name__ == "__main__":
         help="The huggingface repo id for the Llama2 model to be downloaded"
         ", or the path to the huggingface checkpoint folder",
     )
+    parser.add_argument("--lowbit-path", type=str,
+        default="",
+        help="The path to the lowbit model folder, leave blank if you do not want to save. \
+            If path not exists, lowbit model will be saved there. \
+            Else, lowbit model will be loaded.",
+    )
     parser.add_argument('--prompt', type=str, default="What is AI?",
                         help='Prompt to infer')
     parser.add_argument("--n-predict", type=int, default=32, help="Max tokens to predict")
@@ -48,23 +54,38 @@ if __name__ == "__main__":
 
     args = parser.parse_args()
     model_path = args.repo_id_or_model_path
-
-    model = AutoModelForCausalLM.from_pretrained(
-        model_path,
-        torch_dtype=torch.float16,
-        trust_remote_code=True,
-        attn_implementation="eager",
-        load_in_low_bit="sym_int4",
-        optimize_model=True,
-        max_output_len=args.max_output_len,
-        max_prompt_len=args.max_prompt_len,
-        intra_pp=args.intra_pp,
-        inter_pp=args.inter_pp,
-        transpose_value_cache=not args.disable_transpose_value_cache,
-    )
-
+    if not args.lowbit_path or not os.path.exists(args.lowbit_path):
+        model = AutoModelForCausalLM.from_pretrained(
+            model_path,
+            torch_dtype=torch.float16,
+            trust_remote_code=True,
+            attn_implementation="eager",
+            load_in_low_bit="sym_int4",
+            optimize_model=True,
+            max_output_len=args.max_output_len,
+            max_prompt_len=args.max_prompt_len,
+            intra_pp=args.intra_pp,
+            inter_pp=args.inter_pp,
+            transpose_value_cache=not args.disable_transpose_value_cache,
+        )
+    else:
+        model = AutoModelForCausalLM.load_low_bit(
+            args.lowbit_path,
+            attn_implementation="eager",
+            torch_dtype=torch.float16,
+            optimize_model=True,
+            max_output_len=args.max_output_len,
+            max_prompt_len=args.max_prompt_len,
+            intra_pp=args.intra_pp,
+            inter_pp=args.inter_pp,
+            transpose_value_cache=not args.disable_transpose_value_cache,
+            trust_remote_code=True,
+        )
     tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
 
+    if args.lowbit_path and not os.path.exists(args.lowbit_path):
+        model.save_low_bit(args.lowbit_path)
+
     print("-" * 80)
     print("done")
     with torch.inference_mode():
diff --git a/python/llm/src/ipex_llm/transformers/npu_model.py b/python/llm/src/ipex_llm/transformers/npu_model.py
index f04f1317..09940348 100644
--- a/python/llm/src/ipex_llm/transformers/npu_model.py
+++ b/python/llm/src/ipex_llm/transformers/npu_model.py
@@ -270,7 +270,7 @@ class _BaseAutoModelClass:
         invalidInputError(
             qtype in ["sym_int8_rtn", "sym_int4_rtn"],
             f"Unknown bigdl_transformers_low_bit value: {qtype},"
-            f" expected: sym_int4, asym_int4, sym_int5, asym_int5 or sym_int8.",
+            f" expected: sym_int8_rtn, sym_int4_rtn. "
         )
 
         has_remote_code = hasattr(config, "auto_map") and cls.HF_Model.__name__ in config.auto_map