[Chore] Remove unused noqas (#31263)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

tests/conftest.py

@@ -410,7 +410,7 @@ class HfRunner:
 
         # don't put this import at the top level
         # it will call torch.cuda.device_count()
-        from transformers import AutoProcessor  # noqa: F401
+        from transformers import AutoProcessor
 
         self.processor = AutoProcessor.from_pretrained(
             model_name,

tests/entrypoints/openai/test_async_tokenization.py

@@ -15,7 +15,7 @@ MODEL_NAME = "Qwen/Qwen2.5-1.5B-Instruct"
 
 
 @pytest.fixture(scope="module")
-def server():  # noqa: F811
+def server():
     args = [
         # use half precision for speed and memory savings in CI environment
         "--dtype",

tests/entrypoints/openai/test_chat.py

@@ -28,7 +28,7 @@ def zephyr_lora_files():
 
 
 @pytest.fixture(scope="module")
-def server(zephyr_lora_files):  # noqa: F811
+def server(zephyr_lora_files):
     args = [
         # use half precision for speed and memory savings in CI environment
         "--dtype",

tests/entrypoints/openai/test_chat_with_tool_reasoning.py

@@ -12,7 +12,7 @@ MODEL_NAME = "Qwen/QwQ-32B"
 
 
 @pytest.fixture(scope="module")
-def server():  # noqa: F811
+def server():
     args = [
         "--max-model-len",
         "8192",

tests/entrypoints/openai/test_completion_with_function_calling.py

@@ -125,7 +125,7 @@ messages = [
 
 
 @pytest.fixture(scope="module")
-def server():  # noqa: F811
+def server():
     args = [
         # use half precision for speed and memory savings in CI environment
         "--dtype",
@@ -212,7 +212,7 @@ async def test_function_tool_use(
 
 
 @pytest.fixture(scope="module")
-def k2_server():  # noqa: F811
+def k2_server():
     args = [
         # use half precision for speed and memory savings in CI environment
         "--dtype",

tests/entrypoints/openai/test_default_mm_loras.py

@@ -23,7 +23,7 @@ ACTIVE_MM_LORA_RESPONSE = "Spoken text: The first words I spoke in the original
 
 
 @pytest.fixture(scope="module")
-def multimodal_server():  # noqa: F811
+def multimodal_server():
     args = [
         # use half precision for speed and memory savings in CI environment
         "--dtype",

tests/entrypoints/openai/test_enable_force_include_usage.py

@@ -8,7 +8,7 @@ from ...utils import RemoteOpenAIServer
 
 
 @pytest.fixture(scope="module")
-def chat_server_with_force_include_usage(request):  # noqa: F811
+def chat_server_with_force_include_usage(request):
     args = [
         # use half precision for speed and memory savings in CI environment
         "--dtype",

tests/entrypoints/openai/test_messages.py

@@ -11,7 +11,7 @@ MODEL_NAME = "Qwen/Qwen3-0.6B"
 
 
 @pytest.fixture(scope="module")
-def server():  # noqa: F811
+def server():
     args = [
         "--max-model-len",
         "2048",

tests/entrypoints/openai/test_return_tokens_as_ids.py

@@ -37,7 +37,7 @@ def default_server_args(qwen3_lora_files):
 
 
 @pytest.fixture(scope="module")
-def server_fixture(request, default_server_args):  # noqa: F811
+def server_fixture(request, default_server_args):
     use_server_flag = request.param
     if use_server_flag:
         args_with_flag = default_server_args + ["--return-tokens-as-token-ids"]

tests/models/multimodal/generation/test_qwen2_vl.py

@@ -267,7 +267,7 @@ def run_embedding_input_test(
     """Inference result should be the same between
     original image/video input and image/video embeddings input.
     """
-    from transformers import AutoProcessor  # noqa: F401
+    from transformers import AutoProcessor
 
     processor = AutoProcessor.from_pretrained(model)
 

tests/v1/kv_connector/unit/test_example_connector.py

@@ -145,7 +145,7 @@ def test_shared_storage_connector_hashes(tmp_path):
 
     # don't put this import at the top level
     # it will call torch.cuda.device_count()
-    from transformers import AutoProcessor  # noqa: F401
+    from transformers import AutoProcessor
 
     # Create processor to handle the chat prompt
     processor = AutoProcessor.from_pretrained(MODEL_NAME)

vllm/distributed/ec_transfer/ec_connector/example_connector.py

@@ -81,10 +81,7 @@ class ECExampleConnector(ECConnectorBase):
         assert encoder_cache is not None
         if metadata is None:
             logger.warning(
-                (
+                "In connector.start_load_caches, but the connector metadata is None"
-                    "In connector.start_load_caches, ",
-                    "but the connector metadata is None",
-                )
             )
             return
         # Load the EC for each mm data

vllm/entrypoints/serve/elastic_ep/api_router.py

@@ -43,7 +43,7 @@ async def scale_elastic_ep(raw_request: Request):
     try:
         body = await raw_request.json()
     except json.JSONDecodeError as e:
-        raise HTTPException(status_code=400, detail="Invalid JSON format") from e  # noqa: B904
+        raise HTTPException(status_code=400, detail="Invalid JSON format") from e
 
     new_data_parallel_size = body.get("new_data_parallel_size")
     drain_timeout = body.get("drain_timeout", 120)  # Default 2 minutes

vllm/model_executor/model_loader/bitsandbytes_loader.py

@@ -1,6 +1,5 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-# ruff: noqa: SIM117
 import fnmatch
 import glob
 import itertools
@@ -59,7 +58,7 @@ def is_moe_model(model: torch.nn.Module) -> bool:
 
 
 class BitsAndBytesModelLoader(BaseModelLoader):
-    """Model loader to load model weights with BitAndBytes quantization."""
+    """Model loader to load model weights with BitsAndBytes quantization."""
 
     possible_config_file_names = ["adapter_config.json"]
 

vllm/model_executor/model_loader/runai_streamer_loader.py

@@ -1,6 +1,5 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-# ruff: noqa: SIM117
 import os
 from collections.abc import Generator
 

vllm/v1/worker/ec_connector_model_runner_mixin.py

@@ -6,9 +6,7 @@ Define EC connector functionality mixin for model runners.
 
 from collections.abc import Generator
 from contextlib import AbstractContextManager, contextmanager, nullcontext
-from typing import (
+from typing import TYPE_CHECKING
-    TYPE_CHECKING,  # noqa: UP035
-)
 
 import torch
 

vllm/v1/worker/kv_connector_model_runner_mixin.py

@@ -7,9 +7,7 @@ Define KV connector functionality mixin for model runners.
 import copy
 from collections.abc import Generator
 from contextlib import AbstractContextManager, contextmanager, nullcontext
-from typing import (
+from typing import TYPE_CHECKING
-    TYPE_CHECKING,  # noqa: UP035
-)
 
 import torch