diff --git a/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess.py b/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess.py
index 6ef95213..2af272aa 100644
--- a/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess.py
+++ b/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess.py
@@ -8,7 +8,9 @@ from vllm_ascend.utils import enable_custom_op
 
 enable_custom_op()
 
-
+@pytest.mark.skip(    
+    reason="Failure of an individual operator use case causes failures of other operators."
+)
 @pytest.mark.parametrize("cache_mode", ["krope_ctkv", "nzcache"])
 @torch.inference_mode()
 def test_mla_preprocess_kernel(cache_mode: str):
diff --git a/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_nq.py b/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_nq.py
index b18c63f6..907f546b 100644
--- a/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_nq.py
+++ b/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_nq.py
@@ -1,5 +1,6 @@
 import gc
 
+import pytest
 import torch
 import torch_npu
 
@@ -7,7 +8,9 @@ from vllm_ascend.utils import enable_custom_op
 
 enable_custom_op()
 
-
+@pytest.mark.skip(    
+    reason="Failure of an individual operator use case causes failures of other operators."
+)
 @torch.inference_mode()
 def test_mla_preprocess_kernel():
     token_num = 1
diff --git a/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_qdown.py b/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_qdown.py
index 04753617..af414090 100644
--- a/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_qdown.py
+++ b/tests/e2e/nightly/single_node/ops/singlecard_ops/test_mla_preprocess_qdown.py
@@ -8,7 +8,9 @@ from vllm_ascend.utils import enable_custom_op
 
 enable_custom_op()
 
-
+@pytest.mark.skip(    
+    reason="Failure of an individual operator use case causes failures of other operators."
+)
 @pytest.mark.parametrize("cache_mode", ["krope_ctkv", "nzcache"])
 @torch.inference_mode()
 def test_mla_preprocess_kernel(cache_mode: str):