[Frontend][1/n] Improve pooling entrypoints | classify. (#35604)

Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Signed-off-by: wang.yuqi <noooop@126.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
This commit is contained in:
wang.yuqi
2026-03-03 06:05:36 -08:00
committed by GitHub
co-authored by Cyrus Leung
parent 440f0e7dc6
commit ea463978bb
12 changed files with 889 additions and 170 deletions
@@ -0,0 +1,31 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from vllm.config import ModelConfig
from vllm.entrypoints.chat_utils import ChatTemplateConfig
from vllm.entrypoints.pooling.base.io_processor import PoolingIOProcessor
from vllm.renderers import BaseRenderer
from vllm.tasks import SupportedTask
def init_pooling_io_processors(
supported_tasks: tuple[SupportedTask, ...],
model_config: ModelConfig,
renderer: BaseRenderer,
chat_template_config: ChatTemplateConfig,
) -> dict[str, PoolingIOProcessor]:
pooling_io_processors: dict[str, PoolingIOProcessor] = {}
if "classify" in supported_tasks:
from vllm.entrypoints.pooling.classify.io_processor import (
ClassifyIOProcessor,
)
pooling_io_processors["classify"] = ClassifyIOProcessor(
model_config=model_config,
renderer=renderer,
chat_template_config=chat_template_config,
)
return pooling_io_processors