diff --git a/setup.py b/setup.py
index 51ca5e2abecf7..4a20e49235ac8 100644
--- a/setup.py
+++ b/setup.py
@@ -382,8 +382,7 @@ def get_gaudi_sw_version():
     output = subprocess.run("hl-smi",
                             shell=True,
                             text=True,
-                            stdout=subprocess.PIPE,
-                            stderr=subprocess.PIPE,
+                            capture_output=True,
                             env={"ENABLE_CONSOLE": "true"})
     if output.returncode == 0 and output.stdout:
         return output.stdout.split("\n")[2].replace(
diff --git a/vllm/executor/ray_hpu_executor.py b/vllm/executor/ray_hpu_executor.py
index 28d1882cb0db7..a24bab6df370e 100644
--- a/vllm/executor/ray_hpu_executor.py
+++ b/vllm/executor/ray_hpu_executor.py
@@ -34,7 +34,7 @@ class RayHPUExecutor(DistributedGPUExecutor):
     uses_ray: bool = True
 
     def _init_executor(self) -> None:
-        self.forward_dag: Optional["ray.dag.CompiledDAG"] = None
+        self.forward_dag: Optional[ray.dag.CompiledDAG] = None
         # If the env var is set, it uses the Ray's compiled DAG API
         # which optimizes the control plane overhead.
         # Run vLLM with VLLM_USE_RAY_COMPILED_DAG=1 to enable it.
diff --git a/vllm/worker/hpu_model_runner.py b/vllm/worker/hpu_model_runner.py
index 5008a2abd22ea..7e9b2bd13b48a 100644
--- a/vllm/worker/hpu_model_runner.py
+++ b/vllm/worker/hpu_model_runner.py
@@ -67,8 +67,7 @@ class Singleton(type):
 
     def __call__(cls, *args, **kwargs):
         if cls not in cls._instances:
-            cls._instances[cls] = super(Singleton,
-                                        cls).__call__(*args, **kwargs)
+            cls._instances[cls] = super().__call__(*args, **kwargs)
         return cls._instances[cls]
 
 
@@ -273,7 +272,7 @@ def precompute_indices_and_offsets(block_size, slot_mapping, is_prompt):
     return indices, offsets
 
 
-class HpuModelAdapter():
+class HpuModelAdapter:
 
     def __init__(self, model, block_size, dtype, enforce_eager):
         self.model = model
@@ -1643,7 +1642,7 @@ def _maybe_wrap_in_hpu_graph(*args, **kwargs):
     ) if htorch.utils.internal.is_lazy() else HpuModelAdapter(*args, **kwargs)
 
 
-class HabanaProfilerCounterHelper():
+class HabanaProfilerCounterHelper:
 
     def __init__(self):
         self.niter = 0