diff --git a/vllm/entrypoints/grpc/grpc_server.py b/vllm/entrypoints/grpc/grpc_server.py
index 8b5674ed0..662da2faf 100644
--- a/vllm/entrypoints/grpc/grpc_server.py
+++ b/vllm/entrypoints/grpc/grpc_server.py
@@ -15,19 +15,12 @@
                   SamplingParams)
 from vllm.config import ModelConfig
 from vllm.entrypoints.grpc.pb import generation_pb2_grpc
-from vllm.entrypoints.grpc.pb.generation_pb2 import (BatchedGenerationRequest,
-                                                     BatchedGenerationResponse,
-                                                     BatchedTokenizeRequest,
-                                                     BatchedTokenizeResponse,
-                                                     DecodingMethod,
-                                                     GenerationResponse,
-                                                     ModelInfoRequest,
-                                                     ModelInfoResponse,
-                                                     Parameters,
-                                                     ResponseOptions,
-                                                     SingleGenerationRequest,
-                                                     StopReason, TokenInfo,
-                                                     TokenizeResponse)
+from vllm.entrypoints.grpc.pb.generation_pb2 import (
+    BatchedGenerationRequest, BatchedGenerationResponse,
+    BatchedTokenizeRequest, BatchedTokenizeResponse, DecodingMethod,
+    GenerationResponse, ModelInfoRequest, ModelInfoResponse, Parameters,
+    ResponseOptions, SingleGenerationRequest, StopReason, TokenInfo,
+    TokenizeResponse)
 from vllm.entrypoints.grpc.validation import validate_input, validate_params
 from vllm.entrypoints.openai.serving_completion import merge_async_iterators
 from vllm.logger import init_logger
@@ -314,8 +307,7 @@ async def _validate_and_convert_params(
 
         if not greedy and 0.0 < sampling.typical_p < 1.0:
             logits_processors.append(
-                TypicalLogitsWarperWrapper(mass=sampling.typical_p)
-            )
+                TypicalLogitsWarperWrapper(mass=sampling.typical_p))
         if params.decoding.length_penalty is not None:
             length_penalty = (
                 params.decoding.length_penalty.start_index,
@@ -323,7 +315,7 @@ async def _validate_and_convert_params(
             )
             logits_processors.append(
                 LengthPenaltyWarper(length_penalty=length_penalty,
-                    eos_token_id=self.tokenizer.eos_token_id))
+                                    eos_token_id=self.tokenizer.eos_token_id))
 
         time_limit_millis = stopping.time_limit_millis
         deadline = time.time(