[Deprecation] Advance deprecation status (#29617)

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

vllm/config/scheduler.py

@@ -7,7 +7,7 @@ from typing import TYPE_CHECKING, Any, ClassVar, Literal, cast
 
 from pydantic import Field, field_validator
 from pydantic.dataclasses import dataclass
-from typing_extensions import Self, deprecated
+from typing_extensions import Self
 
 from vllm.config.utils import config
 from vllm.logger import init_logger
@@ -224,19 +224,6 @@ class SchedulerConfig:
 
         self.verify_max_model_len(max_model_len)
 
-    @property
-    @deprecated(
-        "`SchedulerConfig.chunked_prefill_enabled` has been renamed to "
-        "`SchedulerConfig.enable_chunked_prefill`. "
-        "The old name will be removed in v0.12."
-    )
-    def chunked_prefill_enabled(self) -> bool:
-        return self.enable_chunked_prefill
-
-    @chunked_prefill_enabled.setter
-    def chunked_prefill_enabled(self, value: bool):
-        self.enable_chunked_prefill = value
-
     def verify_max_model_len(self, max_model_len: int) -> Self:
         if (
             self.max_num_batched_tokens < max_model_len

vllm/distributed/parallel_state.py

@@ -41,7 +41,6 @@ import torch.distributed
 import torch.distributed._functional_collectives as funcol
 import torch.distributed._symmetric_memory
 from torch.distributed import Backend, ProcessGroup
-from typing_extensions import deprecated
 
 import vllm.envs as envs
 from vllm.distributed.device_communicators.base_device_communicator import (
@@ -1078,15 +1077,6 @@ def get_tp_group() -> GroupCoordinator:
     return _TP
 
 
-@deprecated(
-    "`get_tensor_model_parallel_group` has been replaced with "
-    "`get_tp_group` and may be removed after v0.12. Please use "
-    "`get_tp_group` instead."
-)
-def get_tensor_model_parallel_group():
-    return get_tp_group()
-
-
 _DCP: GroupCoordinator | None = None
 
 
@@ -1130,15 +1120,6 @@ def get_pcp_group() -> GroupCoordinator:
     return _PCP
 
 
-@deprecated(
-    "`get_pipeline_model_parallel_group` has been replaced with "
-    "`get_pp_group` and may be removed in v0.12. Please use "
-    "`get_pp_group` instead."
-)
-def get_pipeline_model_parallel_group():
-    return get_pp_group()
-
-
 @contextmanager
 def graph_capture(device: torch.device):
     """

vllm/model_executor/models/utils.py

@@ -10,7 +10,6 @@ import torch
 import torch.nn as nn
 from torch.func import functional_call
 from transformers import PretrainedConfig
-from typing_extensions import deprecated
 
 from vllm.config import VllmConfig
 from vllm.distributed import (
@@ -481,54 +480,6 @@ def _merge_multimodal_embeddings(
     return inputs_embeds
 
 
-@deprecated(
-    "`merge_multimodal_embeddings` has been replaced with "
-    "`SupportsMultiModal.embed_input_ids` and will be "
-    "removed in v0.12."
-)
-def merge_multimodal_embeddings(
-    input_ids: torch.Tensor,
-    inputs_embeds: torch.Tensor,
-    multimodal_embeddings: NestedTensors,
-    placeholder_token_id: int | list[int],
-) -> torch.Tensor:
-    """
-    Merge `multimodal_embeddings` into `inputs_embeds` by overwriting the
-    positions in `inputs_embeds` corresponding to placeholder tokens in
-    `input_ids`.
-
-    `placeholder_token_id` can be a list of token ids (e.g, token ids
-    of img_start, img_break, and img_end tokens) when needed: This means
-    the order of these tokens in the `input_ids` MUST MATCH the order of
-    their embeddings in `multimodal_embeddings` since we need to
-    slice-merge instead of individually scattering.
-
-    For example, if input_ids is "TTTTTSIIIBIIIBIIIETTT", where
-    - T is text token
-    - S is image start token
-    - I is image embedding token
-    - B is image break token
-    - E is image end token.
-
-    Then the image embeddings (that correspond to I's) from vision encoder
-    must be padded with embeddings of S, B, and E in the same order of
-    input_ids for a correct embedding merge.
-
-    Note:
-        This updates `inputs_embeds` in place.
-    """
-    if isinstance(placeholder_token_id, list):
-        is_multimodal = isin_list(input_ids, placeholder_token_id)
-    else:
-        is_multimodal = input_ids == placeholder_token_id
-
-    return _merge_multimodal_embeddings(
-        inputs_embeds,
-        multimodal_embeddings=multimodal_embeddings,
-        is_multimodal=is_multimodal,
-    )
-
-
 def isin_list(
     elements: torch.Tensor,
     test_elements_list: list[int],

vllm/v1/core/sched/output.py

@@ -126,12 +126,12 @@ class CachedRequestData:
         return len(self.req_ids)
 
     @cached_property
-    @deprecated("use resumed_req_ids field")
+    @deprecated("This will be removed in v0.14, use `resumed_req_ids` instead.")
     def resumed_from_preemption(self) -> list[bool]:
         return [req_id in self.resumed_req_ids for req_id in self.req_ids]
 
     @cached_property
-    @deprecated("use all_token_ids field")
+    @deprecated("This will be removed in v0.14, use `all_token_ids` instead.")
     def resumed_req_token_ids(self) -> list[list[int] | None]:
         return [
             self.all_token_ids[req_id] if req_id in self.resumed_req_ids else None