[Frontend] Consolidate online serving utils. (#44479)

Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
This commit is contained in:
wang.yuqi
2026-06-04 06:49:31 +00:00
committed by GitHub
parent b4b4aaa70e
commit d01d0b4646
81 changed files with 466 additions and 435 deletions
+15 -15
View File
@@ -27,12 +27,22 @@ from vllm.engine.arg_utils import AsyncEngineArgs
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.chat_utils import load_chat_template
from vllm.entrypoints.launcher import serve_http
from vllm.entrypoints.logger import RequestLogger
from vllm.entrypoints.openai.cli_args import make_arg_parser, validate_parsed_serve_args
from vllm.entrypoints.openai.engine.protocol import GenerationError
from vllm.entrypoints.openai.models.protocol import BaseModelPath
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.openai.server_utils import (
from vllm.entrypoints.serve.elastic_ep.middleware import ScalingMiddleware
from vllm.entrypoints.serve.render.serving import OpenAIServingRender
from vllm.entrypoints.serve.sagemaker.api_router import sagemaker_standards_bootstrap
from vllm.entrypoints.serve.tokenize.serving import OpenAIServingTokenization
from vllm.entrypoints.serve.utils.api_utils import (
cli_env_setup,
log_non_default_args,
log_version_and_model,
process_lora_modules,
)
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.entrypoints.serve.utils.server_utils import (
engine_error_handler,
exception_handler,
generation_error_handler,
@@ -42,16 +52,6 @@ from vllm.entrypoints.openai.server_utils import (
log_response,
validation_exception_handler,
)
from vllm.entrypoints.sagemaker.api_router import sagemaker_standards_bootstrap
from vllm.entrypoints.serve.elastic_ep.middleware import ScalingMiddleware
from vllm.entrypoints.serve.render.serving import OpenAIServingRender
from vllm.entrypoints.serve.tokenize.serving import OpenAIServingTokenization
from vllm.entrypoints.utils import (
cli_env_setup,
log_non_default_args,
log_version_and_model,
process_lora_modules,
)
from vllm.logger import init_logger
from vllm.reasoning import ReasoningParserManager
from vllm.tasks import POOLING_TASKS, SupportedTask
@@ -187,7 +187,7 @@ def build_app(
register_models_api_router(app)
from vllm.entrypoints.sagemaker.api_router import (
from vllm.entrypoints.serve.sagemaker.api_router import (
attach_router as register_sagemaker_api_router,
)
@@ -254,12 +254,12 @@ def build_app(
# Ensure --api-key option from CLI takes precedence over VLLM_API_KEY
if tokens := [key for key in (args.api_key or [envs.VLLM_API_KEY]) if key]:
from vllm.entrypoints.openai.server_utils import AuthenticationMiddleware
from vllm.entrypoints.serve.utils.server_utils import AuthenticationMiddleware
app.add_middleware(AuthenticationMiddleware, tokens=tokens)
if args.enable_request_id_headers:
from vllm.entrypoints.openai.server_utils import XRequestIdMiddleware
from vllm.entrypoints.serve.utils.server_utils import XRequestIdMiddleware
app.add_middleware(XRequestIdMiddleware)