diff --git a/examples/offline_inference/data_parallel.py b/examples/offline_inference/data_parallel.py
index a9544c8cf..2e1fa50e2 100644
--- a/examples/offline_inference/data_parallel.py
+++ b/examples/offline_inference/data_parallel.py
@@ -48,15 +48,16 @@ def main(dp_size, dp_rank, dp_master_ip, dp_master_port, GPUs_per_dp_rank):
                                      max_tokens=16 * (dp_rank + 1))
 
     # Create an LLM.
-    llm = LLM(model="facebook/opt-125m", tensor_parallel_size=2, enforce_eager=True)
+    llm = LLM(model="facebook/opt-125m",
+              tensor_parallel_size=2,
+              enforce_eager=True)
     outputs = llm.generate(prompts, sampling_params)
     # Print the outputs.
     for output in outputs:
         prompt = output.prompt
         generated_text = output.outputs[0].text
-        print(
-            f"DP rank {dp_rank}, Prompt: {prompt!r}, "
-            f"Generated text: {generated_text!r}")
+        print(f"DP rank {dp_rank}, Prompt: {prompt!r}, "
+              f"Generated text: {generated_text!r}")
 
 
 if __name__ == "__main__":
diff --git a/vllm/config.py b/vllm/config.py
index ed32a5028..d3139b5fd 100644
--- a/vllm/config.py
+++ b/vllm/config.py
@@ -1372,7 +1372,7 @@ class ParallelConfig:
 
     @staticmethod
     def has_unfinished_dp(dp_group: "ProcessGroup",
-                                      has_unfinished: bool) -> bool:
+                          has_unfinished: bool) -> bool:
         tensor = torch.tensor([has_unfinished],
                               dtype=torch.int32,
                               device="cpu")
diff --git a/vllm/utils.py b/vllm/utils.py
index 7d2415492..675edc362 100644
--- a/vllm/utils.py
+++ b/vllm/utils.py
@@ -518,6 +518,7 @@ def get_open_port() -> int:
             return port
     return _get_open_port()
 
+
 def _get_open_port() -> int:
     port = envs.VLLM_PORT
     if port is not None:
diff --git a/vllm/v1/engine/core_client.py b/vllm/v1/engine/core_client.py
index e898a872c..527aa7283 100644
--- a/vllm/v1/engine/core_client.py
+++ b/vllm/v1/engine/core_client.py
@@ -89,7 +89,7 @@ class EngineCoreClient(ABC):
 
     def execute_dummy_batch(self) -> None:
         raise NotImplementedError
-    
+
     async def execute_dummy_batch_async(self) -> None:
         raise NotImplementedError
 
@@ -343,6 +343,7 @@ class SyncMPClient(MPClient):
     def execute_dummy_batch(self) -> None:
         self._call_utility("execute_dummy_batch")
 
+
 class AsyncMPClient(MPClient):
     """Asyncio-compatible client for multi-proc EngineCore."""
 
diff --git a/vllm/v1/worker/gpu_model_runner.py b/vllm/v1/worker/gpu_model_runner.py
index f002cbfcc..a7b9d4781 100644
--- a/vllm/v1/worker/gpu_model_runner.py
+++ b/vllm/v1/worker/gpu_model_runner.py
@@ -1167,7 +1167,8 @@ class GPUModelRunner(LoRAModelRunnerMixin):
                 for k, v in self.intermediate_tensors.items()
             })
 
-        with set_forward_context(None, self.vllm_config, num_tokens=num_tokens):
+        with set_forward_context(None, self.vllm_config,
+                                 num_tokens=num_tokens):
             hidden_states = model(
                 input_ids=input_ids,
                 positions=positions,