diff --git a/.github/workflows/unit-test-optimize.yml b/.github/workflows/unit-test-optimize.yml
index 6399df03878..edfad46b42b 100644
--- a/.github/workflows/unit-test-optimize.yml
+++ b/.github/workflows/unit-test-optimize.yml
@@ -45,7 +45,7 @@ jobs:
             test_name: "PR-test"
           - test_branch: "main"
             test_name: "baseline"
-      fail-fast: true
+      fail-fast: false
     name: optimize-unit-test-${{ matrix.test_name }}
     steps:
       - name: Docker Clean Up
diff --git a/intel_extension_for_transformers/transformers/modeling/modeling_auto.py b/intel_extension_for_transformers/transformers/modeling/modeling_auto.py
index d54837e2220..bf8420f6b8f 100644
--- a/intel_extension_for_transformers/transformers/modeling/modeling_auto.py
+++ b/intel_extension_for_transformers/transformers/modeling/modeling_auto.py
@@ -67,7 +67,11 @@
     convert_to_smoothquant_model,
     replace_linear,
 )
-from ...tools.utils import is_intel_gpu_available, is_ipex_available, _neural_compressor_version
+from ...tools.utils import (
+    is_intel_gpu_available,
+    is_ipex_available,
+    _neural_compressor_version,
+)
 from accelerate import init_empty_weights
 from huggingface_hub import hf_hub_download
 from neural_compressor.torch.algorithms.weight_only.modules import WeightOnlyLinear
@@ -1832,7 +1836,6 @@ 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:
@@ -1949,7 +1952,6 @@ 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,