A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[Misc] Remove `isort` and `yapf` ignores (#26888)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Cyrus Leung committed
Oct 15, 2025 at 20:09 UTC
f93e348010d202297825efb25fa4138913bdfea4
4 files changed
+2
-10
tests/v1/engine/conftest.py
+1
-2
@@ -6,6 +6,7 @@ import torch
6
from transformers import AutoTokenizer
7
8
from tests.v1.engine.utils import (
9
+ FULL_STRINGS,
10
NUM_PROMPT_LOGPROBS_UNDER_TEST,
11
NUM_SAMPLE_LOGPROBS_UNDER_TEST,
12
PROMPT_LEN,
@@ -18,8 +19,6 @@ from vllm.engine.arg_utils import EngineArgs
19
20
from ...distributed.conftest import publisher_config, random_port # noqa: F401
21
21
-from tests.v1.engine.utils import FULL_STRINGS # isort: skip
22
-
22
EngineCoreSampleLogprobsType = list[tuple[torch.Tensor, torch.Tensor]]
23
EngineCorePromptLogprobsType = tuple[torch.Tensor, torch.Tensor]
24
tests/v1/tpu/test_topk_topp_sampler.py
-3
@@ -8,10 +8,7 @@ import torch_xla
8
9
from vllm.platforms import current_platform
10
from vllm.v1.sample.ops.topk_topp_sampler import apply_top_k_top_p
11
-
12
-# isort: off
11
from vllm.v1.sample.tpu.sampler import apply_top_k_top_p as apply_top_k_top_p_tpu
14
-# isort: on
12
13
if not current_platform.is_tpu():
14
pytest.skip("This test needs a TPU.", allow_module_level=True)
vllm/model_executor/layers/quantization/compressed_tensors/schemes/__init__.py
+1
@@ -15,6 +15,7 @@ from .compressed_tensors_w8a8_int8 import CompressedTensorsW8A8Int8
15
from .compressed_tensors_w8a16_fp8 import CompressedTensorsW8A16Fp8
16
from .compressed_tensors_wNa16 import WNA16_SUPPORTED_BITS, CompressedTensorsWNA16
17
18
+# This avoids circular import error
19
from .compressed_tensors_24 import CompressedTensors24 # isort: skip
20
21
__all__ = [
vllm/model_executor/models/qwen3_omni_moe_thinker.py
-5
@@ -80,17 +80,12 @@ from .interfaces import (
80
SupportsMultiModal,
81
SupportsPP,
82
)
83
-
84
-# yapf conflicts with isort for this block
85
-# yapf: disable
83
from .qwen2_5_omni_thinker import (
84
Qwen2_5OmniConditionalGenerationMixin,
85
Qwen2_5OmniThinkerDummyInputsBuilder,
86
Qwen2_5OmniThinkerMultiModalProcessor,
87
Qwen2_5OmniThinkerProcessingInfo,
88
)
92
-
93
-# yapf: enable
89
from .qwen2_5_vl import (
90
Qwen2_5_VisionAttention,
91
Qwen2_5_VisionRotaryEmbedding,