diff --git a/src/accelerate/test_utils/__init__.py b/src/accelerate/test_utils/__init__.py
index 0dbee933ed3..36e78d15487 100644
--- a/src/accelerate/test_utils/__init__.py
+++ b/src/accelerate/test_utils/__init__.py
@@ -13,6 +13,8 @@
     require_multi_gpu,
     require_multi_xpu,
     require_non_cpu,
+    require_non_xpu,
+    require_npu,
     require_pippy,
     require_single_device,
     require_single_gpu,
diff --git a/src/accelerate/test_utils/testing.py b/src/accelerate/test_utils/testing.py
index cdeee5fa995..b7b3e088ea4 100644
--- a/src/accelerate/test_utils/testing.py
+++ b/src/accelerate/test_utils/testing.py
@@ -129,6 +129,20 @@ def require_xpu(test_case):
     return unittest.skipUnless(is_xpu_available(), "test requires a XPU")(test_case)
 
 
+def require_non_xpu(test_case):
+    """
+    Decorator marking a test that should be skipped for XPU.
+    """
+    return unittest.skipUnless(torch_device != "xpu", "test requires a non-XPU")(test_case)
+
+
+def require_npu(test_case):
+    """
+    Decorator marking a test that requires NPU. These tests are skipped when there are no NPU available.
+    """
+    return unittest.skipUnless(is_npu_available(), "test require a NPU")(test_case)
+
+
 def require_mps(test_case):
     """
     Decorator marking a test that requires MPS backend. These tests are skipped when torch doesn't support `mps`
diff --git a/tests/test_kwargs_handlers.py b/tests/test_kwargs_handlers.py
index ce817785a55..28c12b79a4c 100644
--- a/tests/test_kwargs_handlers.py
+++ b/tests/test_kwargs_handlers.py
@@ -21,7 +21,13 @@
 
 from accelerate import Accelerator, DistributedDataParallelKwargs, GradScalerKwargs
 from accelerate.state import AcceleratorState
-from accelerate.test_utils import device_count, execute_subprocess_async, require_multi_device, require_non_cpu
+from accelerate.test_utils import (
+    device_count,
+    execute_subprocess_async,
+    require_multi_device,
+    require_non_cpu,
+    require_non_xpu,
+)
 from accelerate.utils import AutocastKwargs, KwargsHandler, TorchDynamoPlugin, clear_environment
 
 
@@ -41,6 +47,7 @@ def test_kwargs_handler(self):
         self.assertDictEqual(MockClass(a=2, c=2.25).to_kwargs(), {"a": 2, "c": 2.25})
 
     @require_non_cpu
+    @require_non_xpu
     def test_grad_scaler_kwargs(self):
         # If no defaults are changed, `to_kwargs` returns an empty dict.
         scaler_handler = GradScalerKwargs(init_scale=1024, growth_factor=2)
diff --git a/tests/test_optimizer.py b/tests/test_optimizer.py
index ff67ec0979e..ad3210ecdb9 100644
--- a/tests/test_optimizer.py
+++ b/tests/test_optimizer.py
@@ -19,7 +19,7 @@
 
 from accelerate import Accelerator
 from accelerate.state import AcceleratorState
-from accelerate.test_utils import require_cpu, require_non_cpu
+from accelerate.test_utils import require_cpu, require_non_cpu, require_non_xpu
 
 
 @require_cpu
@@ -37,6 +37,7 @@ def test_accelerated_optimizer_pickling(self):
 
 
 @require_non_cpu
+@require_non_xpu
 class OptimizerTester(unittest.TestCase):
     def test_accelerated_optimizer_step_was_skipped(self):
         model = torch.nn.Linear(5, 5)