[Frontend][4/n] Improve pooling entrypoints | pooling. (#39153)

Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
This commit is contained in:
wang.yuqi
2026-04-09 10:09:45 +00:00
committed by GitHub
parent b6c9be509e
commit 66c079ae83
43 changed files with 554 additions and 733 deletions
@@ -86,7 +86,6 @@ def _build_serving_tokens(engine: AsyncLLM, **kwargs) -> ServingTokens:
serving_render = OpenAIServingRender(
model_config=engine.model_config,
renderer=engine.renderer,
io_processor=engine.io_processor,
model_registry=models.registry,
request_logger=None,
chat_template=None,
@@ -148,7 +147,6 @@ def _mock_engine() -> MagicMock:
engine.errored = False
engine.model_config = MockModelConfig()
engine.input_processor = MagicMock()
engine.io_processor = MagicMock()
engine.renderer = _build_renderer(engine.model_config)
return engine
@@ -34,7 +34,6 @@ async def _async_serving_models_init() -> OpenAIServingModels:
mock_model_config.max_model_len = 2048
mock_engine_client.model_config = mock_model_config
mock_engine_client.input_processor = MagicMock()
mock_engine_client.io_processor = MagicMock()
mock_engine_client.renderer = MagicMock()
serving_models = OpenAIServingModels(