[Misc] Move utils to avoid conflicts with stdlib, and move tests (#27169)

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

vllm/entrypoints/openai/serving_pooling.py

@@ -36,7 +36,7 @@ from vllm.entrypoints.utils import _validate_truncation_size
 from vllm.logger import init_logger
 from vllm.outputs import PoolingOutput, PoolingRequestOutput
 from vllm.tasks import SupportedTask
-from vllm.utils.asyncio import merge_async_iterators
+from vllm.utils.async_utils import merge_async_iterators
 
 logger = init_logger(__name__)
 

vllm/entrypoints/openai/serving_score.py

@@ -37,7 +37,7 @@ from vllm.logger import init_logger
 from vllm.lora.request import LoRARequest
 from vllm.outputs import PoolingRequestOutput, ScoringRequestOutput
 from vllm.transformers_utils.tokenizer import AnyTokenizer, MistralTokenizer
-from vllm.utils.asyncio import make_async, merge_async_iterators
+from vllm.utils.async_utils import make_async, merge_async_iterators
 
 logger = init_logger(__name__)
 

vllm/entrypoints/openai/tool_parsers/abstract_tool_parser.py

@@ -12,7 +12,7 @@ from vllm.entrypoints.openai.protocol import (
 )
 from vllm.logger import init_logger
 from vllm.transformers_utils.tokenizer import AnyTokenizer
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 from vllm.utils.import_utils import import_from_path
 
 logger = init_logger(__name__)

vllm/entrypoints/renderer.py

@@ -17,7 +17,7 @@ from vllm.inputs.data import TextPrompt as EngineTextPrompt
 from vllm.inputs.data import TokensPrompt as EngineTokensPrompt
 from vllm.inputs.parse import get_prompt_components, parse_raw_prompts
 from vllm.transformers_utils.tokenizer import AnyTokenizer
-from vllm.utils.asyncio import AsyncMicrobatchTokenizer
+from vllm.utils.async_utils import AsyncMicrobatchTokenizer
 
 
 @dataclass(frozen=True)

vllm/executor/executor_base.py

@@ -17,7 +17,7 @@ from vllm.logger import init_logger
 from vllm.lora.request import LoRARequest
 from vllm.sequence import ExecuteModelRequest
 from vllm.tasks import SupportedTask
-from vllm.utils.asyncio import make_async
+from vllm.utils.async_utils import make_async
 from vllm.v1.outputs import SamplerOutput
 from vllm.v1.worker.worker_base import WorkerBase
 

vllm/executor/ray_distributed_executor.py

@@ -19,7 +19,7 @@ from vllm.logger import init_logger
 from vllm.platforms import current_platform
 from vllm.ray.ray_env import get_env_vars_to_copy
 from vllm.sequence import ExecuteModelRequest
-from vllm.utils.asyncio import make_async
+from vllm.utils.async_utils import make_async
 from vllm.utils.network_utils import (
     get_distributed_init_method,
     get_ip,

vllm/inputs/parse.py

@@ -5,7 +5,7 @@ from typing import TYPE_CHECKING, Literal, NamedTuple, TypeAlias, TypedDict, cas
 
 from typing_extensions import TypeIs
 
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 
 from .data import (
     EmbedsPrompt,

vllm/model_executor/layers/activation.py

@@ -17,7 +17,7 @@ from vllm.logger import init_logger
 from vllm.model_executor.custom_op import CustomOp
 from vllm.model_executor.utils import set_weight_attrs
 from vllm.platforms import current_platform
-from vllm.utils.collections import LazyDict
+from vllm.utils.collection_utils import LazyDict
 
 logger = init_logger(__name__)
 

vllm/model_executor/layers/fused_moe/deep_gemm_moe.py

@@ -31,7 +31,7 @@ from vllm.utils.deep_gemm import (
     get_mk_alignment_for_contiguous_layout,
     m_grouped_fp8_gemm_nt_contiguous,
 )
-from vllm.utils.functools import run_once
+from vllm.utils.func_utils import run_once
 
 logger = init_logger(__name__)
 

vllm/model_executor/layers/quantization/gptq.py

@@ -28,7 +28,7 @@ from vllm.model_executor.parameter import (
     RowvLLMParameter,
 )
 from vllm.transformers_utils.config import get_safetensors_params_metadata
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 
 if TYPE_CHECKING:
     from vllm.model_executor.layers.quantization import QuantizationMethods

vllm/model_executor/layers/quantization/gptq_marlin.py

@@ -57,7 +57,7 @@ from vllm.model_executor.parameter import (
 from vllm.platforms import current_platform
 from vllm.scalar_type import scalar_types
 from vllm.transformers_utils.config import get_safetensors_params_metadata
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 
 logger = init_logger(__name__)
 

vllm/model_executor/models/deepseek_vl2.py

@@ -48,7 +48,7 @@ from vllm.transformers_utils.configs.deepseek_vl2 import (
 )
 from vllm.transformers_utils.processors.deepseek_vl2 import DeepseekVLV2Processor
 from vllm.transformers_utils.tokenizer import cached_tokenizer_from_config
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 from vllm.utils.tensor_schema import TensorSchema, TensorShape
 from vllm.utils.torch_utils import set_default_torch_dtype
 

vllm/model_executor/models/interfaces.py

@@ -24,7 +24,7 @@ from vllm.inputs import TokensPrompt
 from vllm.inputs.data import PromptType
 from vllm.logger import init_logger
 from vllm.model_executor.layers.quantization import QuantizationConfig
-from vllm.utils.functools import supports_kw
+from vllm.utils.func_utils import supports_kw
 
 from .interfaces_base import VllmModel, is_pooling_model
 

vllm/model_executor/models/interfaces_base.py

@@ -15,7 +15,7 @@ import torch.nn as nn
 from typing_extensions import TypeIs, TypeVar
 
 from vllm.logger import init_logger
-from vllm.utils.functools import supports_kw
+from vllm.utils.func_utils import supports_kw
 
 if TYPE_CHECKING:
     from vllm.config import VllmConfig

vllm/model_executor/models/llava_next_video.py

@@ -33,7 +33,7 @@ from vllm.multimodal.processing import (
 )
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
 from vllm.sequence import IntermediateTensors
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 from vllm.utils.tensor_schema import TensorSchema, TensorShape
 
 from .interfaces import MultiModalEmbeddings, SupportsMultiModal, SupportsPP

vllm/model_executor/models/minicpmv.py

@@ -85,7 +85,7 @@ from vllm.multimodal.processing import (
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
 from vllm.platforms import current_platform
 from vllm.sequence import IntermediateTensors
-from vllm.utils.collections import flatten_2d_lists
+from vllm.utils.collection_utils import flatten_2d_lists
 from vllm.utils.tensor_schema import TensorSchema, TensorShape
 from vllm.utils.torch_utils import set_default_torch_dtype
 

vllm/model_executor/models/qwen3_vl.py

@@ -79,7 +79,7 @@ from vllm.multimodal.processing import (
 )
 from vllm.multimodal.profiling import BaseDummyInputsBuilder
 from vllm.sequence import IntermediateTensors
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 
 from .interfaces import (
     MultiModalEmbeddings,

vllm/multimodal/inputs.py

@@ -22,7 +22,7 @@ from typing import (
 import numpy as np
 from typing_extensions import NotRequired, TypeVar, deprecated
 
-from vllm.utils.collections import full_groupby, is_list_of
+from vllm.utils.collection_utils import full_groupby, is_list_of
 from vllm.utils.import_utils import LazyLoader
 from vllm.utils.jsontree import json_map_leaves
 

vllm/multimodal/parse.py

@@ -19,7 +19,7 @@ import numpy as np
 import torch
 from typing_extensions import assert_never
 
-from vllm.utils.collections import is_list_of
+from vllm.utils.collection_utils import is_list_of
 from vllm.utils.import_utils import LazyLoader
 
 from .audio import AudioResampler

vllm/multimodal/processing.py

@@ -25,8 +25,8 @@ from typing_extensions import TypeVar, assert_never
 from vllm.logger import init_logger
 from vllm.transformers_utils.processor import cached_processor_from_config
 from vllm.transformers_utils.tokenizer import AnyTokenizer, decode_tokens, encode_tokens
-from vllm.utils.collections import flatten_2d_lists, full_groupby
-from vllm.utils.functools import get_allowed_kwarg_only_overrides
+from vllm.utils.collection_utils import flatten_2d_lists, full_groupby
+from vllm.utils.func_utils import get_allowed_kwarg_only_overrides
 from vllm.utils.jsontree import JSONTree, json_map_leaves
 
 from .hasher import MultiModalHasher
@@ -486,7 +486,7 @@ _M = TypeVar("_M", bound=_HasModalityAttr | _HasModalityProp)
 def full_groupby_modality(values: Iterable[_M]) -> ItemsView[str, list[_M]]:
     """
     Convenience function to apply
-    [`full_groupby`][vllm.utils.collections.full_groupby]
+    [`full_groupby`][vllm.utils.collection_utils.full_groupby]
     based on modality.
     """
     return full_groupby(values, key=lambda x: x.modality)