diff --git a/intel_extension_for_transformers/transformers/modeling/modeling_auto.py b/intel_extension_for_transformers/transformers/modeling/modeling_auto.py
index 77c8008b063..63540e11a74 100644
--- a/intel_extension_for_transformers/transformers/modeling/modeling_auto.py
+++ b/intel_extension_for_transformers/transformers/modeling/modeling_auto.py
@@ -1833,7 +1833,6 @@ def load_low_bit(cls, pretrained_model_name_or_path, *model_args, **kwargs):
         if quantization_config.weight_dtype not in [
             "fp8_e5m2",
             "fp8_e4m3",
-            "int4_fullrange"
         ]:
             model = build_woq_model(model, quantization_config)
         else:
@@ -1950,7 +1949,6 @@ def replace_ipex_cpu_woq_linear(model, current_name=[]):
         if quantization_config.weight_dtype not in [
             "fp8_e5m2",
             "fp8_e4m3",
-            "int4_fullrange"
         ] and not quantization_config.use_ipex:
             model = replace_linear(
                 model,