Scheduled removal of `override_pooler_config` and `disable_log_requests` (#29402)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

vllm/config/model.py

@@ -299,9 +299,6 @@ class ModelConfig:
     pooler_config: PoolerConfig | None = None
     """Pooler config which controls the behaviour of output pooling in pooling
     models."""
-    override_pooler_config: dict | PoolerConfig | None = None
-    """[DEPRECATED] Use `pooler_config` instead. This field will be removed in
-    v0.12.0 or v1.0.0, whichever is sooner."""
 
     # Multimodal config and init vars
     multimodal_config: MultiModalConfig | None = None
@@ -359,7 +356,6 @@ class ModelConfig:
             "logits_processors",
             "io_processor_plugin",
             "pooler_config",
-            "override_pooler_config",
             "multimodal_config",
             "limit_mm_per_prompt",
             "media_io_kwargs",
@@ -648,18 +644,6 @@ class ModelConfig:
 
         # Init pooler config if needed
         if self.runner_type == "pooling":
-            if self.override_pooler_config is not None:
-                logger.warning_once(
-                    "`override_pooler_config` is deprecated and will be "
-                    "removed in v0.12.0 or v1.0.0, whichever is sooner. "
-                    "Please use `pooler_config` instead."
-                )
-
-                if isinstance(self.override_pooler_config, dict):
-                    self.pooler_config = PoolerConfig(**self.override_pooler_config)
-                else:
-                    self.pooler_config = self.override_pooler_config
-
             if self.pooler_config is None:
                 self.pooler_config = PoolerConfig()
 

vllm/engine/arg_utils.py

@@ -29,7 +29,7 @@ import regex as re
 import torch
 from pydantic import TypeAdapter, ValidationError
 from pydantic.fields import FieldInfo
-from typing_extensions import TypeIs, deprecated
+from typing_extensions import TypeIs
 
 import vllm.envs as envs
 from vllm.attention.backends.registry import AttentionBackendEnum
@@ -520,9 +520,6 @@ class EngineArgs:
     scheduler_cls: str | type[object] | None = SchedulerConfig.scheduler_cls
 
     pooler_config: PoolerConfig | None = ModelConfig.pooler_config
-    override_pooler_config: dict | PoolerConfig | None = (
-        ModelConfig.override_pooler_config
-    )
     compilation_config: CompilationConfig = get_field(VllmConfig, "compilation_config")
     worker_cls: str = ParallelConfig.worker_cls
     worker_extension_cls: str = ParallelConfig.worker_extension_cls
@@ -659,11 +656,6 @@ class EngineArgs:
         )
         model_group.add_argument("--hf-overrides", **model_kwargs["hf_overrides"])
         model_group.add_argument("--pooler-config", **model_kwargs["pooler_config"])
-        model_group.add_argument(
-            "--override-pooler-config",
-            **model_kwargs["override_pooler_config"],
-            deprecated=True,
-        )
         model_group.add_argument(
             "--logits-processor-pattern", **model_kwargs["logits_processor_pattern"]
         )
@@ -1243,7 +1235,6 @@ class EngineArgs:
             mm_encoder_tp_mode=self.mm_encoder_tp_mode,
             mm_encoder_attn_backend=self.mm_encoder_attn_backend,
             pooler_config=self.pooler_config,
-            override_pooler_config=self.override_pooler_config,
             logits_processor_pattern=self.logits_processor_pattern,
             generation_config=self.generation_config,
             override_generation_config=self.override_generation_config,
@@ -2047,24 +2038,6 @@ class AsyncEngineArgs(EngineArgs):
 
     enable_log_requests: bool = False
 
-    @property
-    @deprecated(
-        "`disable_log_requests` is deprecated and has been replaced with "
-        "`enable_log_requests`. This will be removed in v0.12.0. Please use "
-        "`enable_log_requests` instead."
-    )
-    def disable_log_requests(self) -> bool:
-        return not self.enable_log_requests
-
-    @disable_log_requests.setter
-    @deprecated(
-        "`disable_log_requests` is deprecated and has been replaced with "
-        "`enable_log_requests`. This will be removed in v0.12.0. Please use "
-        "`enable_log_requests` instead."
-    )
-    def disable_log_requests(self, value: bool):
-        self.enable_log_requests = not value
-
     @staticmethod
     def add_cli_args(
         parser: FlexibleArgumentParser, async_args_only: bool = False

vllm/entrypoints/llm.py

@@ -174,9 +174,6 @@ class LLM:
             For example, for Phi-3-Vision: `{"num_crops": 4}`.
         pooler_config: Initialize non-default pooling config for the pooling
             model. e.g. `PoolerConfig(pooling_type="mean", normalize=False)`.
-        override_pooler_config: [DEPRECATED] Use `pooler_config` instead. This
-            argument is deprecated and will be removed in v0.12.0 or v1.0.0,
-            whichever is sooner.
         compilation_config: Either an integer or a dictionary. If it is an
             integer, it is used as the mode of compilation optimization. If it
             is a dictionary, it can specify the full compilation configuration.
@@ -214,7 +211,6 @@ class LLM:
         hf_overrides: HfOverrides | None = None,
         mm_processor_kwargs: dict[str, Any] | None = None,
         pooler_config: PoolerConfig | None = None,
-        override_pooler_config: PoolerConfig | None = None,
         structured_outputs_config: dict[str, Any]
         | StructuredOutputsConfig
         | None = None,
@@ -330,7 +326,6 @@ class LLM:
             hf_overrides=hf_overrides,
             mm_processor_kwargs=mm_processor_kwargs,
             pooler_config=pooler_config,
-            override_pooler_config=override_pooler_config,
             structured_outputs_config=structured_outputs_instance,
             compilation_config=compilation_config_instance,
             logits_processors=logits_processors,

vllm/utils/argparse_utils.py

@@ -73,14 +73,6 @@ class FlexibleArgumentParser(ArgumentParser):
         # Enable the deprecated kwarg for Python 3.12 and below
 
         def parse_known_args(self, args=None, namespace=None):
-            if args is not None and "--disable-log-requests" in args:
-                # Special case warning because the warning below won't trigger
-                # if –-disable-log-requests because its value is default.
-                logger.warning_once(
-                    "argument '--disable-log-requests' is deprecated and "
-                    "replaced with '--enable-log-requests'. This will be "
-                    "removed in v0.12.0."
-                )
             namespace, args = super().parse_known_args(args, namespace)
             for action in FlexibleArgumentParser._deprecated:
                 if (

vllm/v1/engine/async_llm.py

@@ -31,7 +31,6 @@ from vllm.transformers_utils.tokenizer import AnyTokenizer, init_tokenizer_from_
 from vllm.usage.usage_lib import UsageContext
 from vllm.utils.async_utils import cancel_task_threadsafe
 from vllm.utils.collection_utils import as_list
-from vllm.utils.func_utils import deprecate_kwargs
 from vllm.utils.math_utils import cdiv
 from vllm.v1.engine import EngineCoreRequest
 from vllm.v1.engine.core_client import EngineCoreClient
@@ -195,12 +194,6 @@ class AsyncLLM(EngineClient):
             self.profiler = None
 
     @classmethod
-    @deprecate_kwargs(
-        "disable_log_requests",
-        additional_message=(
-            "This argument will have no effect. Use `enable_log_requests` instead."
-        ),
-    )
     def from_vllm_config(
         cls,
         vllm_config: VllmConfig,
@@ -213,7 +206,6 @@ class AsyncLLM(EngineClient):
         client_addresses: dict[str, str] | None = None,
         client_count: int = 1,
         client_index: int = 0,
-        disable_log_requests: bool = True,  # Deprecated, will be removed
     ) -> "AsyncLLM":
         # Create the LLMEngine.
         return cls(