[V0 Deprecation] Remove pooling model support in V0 (#23434)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Maximilien de Bayser
2025-08-29 00:04:02 -07:00
committed by GitHub
co-authored by Woosuk Kwon
parent 934bebf192
commit 2554b27baa
38 changed files with 99 additions and 808 deletions
+2 -12
View File
@@ -13,10 +13,9 @@ from vllm.config import VllmConfig
from vllm.logger import init_logger
from vllm.model_executor.layers.sampler import SamplerOutput
from vllm.model_executor.models.interfaces import supports_transcription
from vllm.model_executor.models.interfaces_base import (
is_pooling_model, is_text_generation_model)
from vllm.model_executor.models.interfaces_base import is_text_generation_model
from vllm.sequence import IntermediateTensors, SequenceGroupMetadata
from vllm.tasks import GenerationTask, PoolingTask, SupportedTask
from vllm.tasks import GenerationTask, SupportedTask
if TYPE_CHECKING:
from vllm.attention import AttentionMetadata
@@ -241,20 +240,11 @@ class ModelRunnerBase(ABC, Generic[T]):
return supported_tasks
def get_supported_pooling_tasks(self) -> list[PoolingTask]:
model = self.get_model()
if not is_pooling_model(model):
return []
return list(model.pooler.get_supported_tasks())
def get_supported_tasks(self) -> tuple[SupportedTask, ...]:
tasks = list[SupportedTask]()
if self.model_config.runner_type == "generate":
tasks.extend(self.get_supported_generation_tasks())
if self.model_config.runner_type == "pooling":
tasks.extend(self.get_supported_pooling_tasks())
return tuple(tasks)