[Frontend] Refactor ServingTokenization entrypoint. (#46022)

Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
This commit is contained in:
wang.yuqi
2026-06-22 06:27:58 +00:00
committed by GitHub
parent 31124749d1
commit 1eb2cc961e
15 changed files with 322 additions and 302 deletions
@@ -5,8 +5,7 @@ from fastapi import APIRouter, Request
from fastapi.responses import JSONResponse
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.serving import OpenAIServing
from vllm.entrypoints.serve.tokenize.serving import OpenAIServingTokenization
from vllm.entrypoints.serve.tokenize.serving import ServingTokenization
from vllm.logger import init_logger
from vllm.version import __version__ as VLLM_VERSION
@@ -15,13 +14,13 @@ router = APIRouter()
logger = init_logger(__name__)
def base(request: Request) -> OpenAIServing:
def base(request: Request) -> ServingTokenization:
# Reuse the existing instance
return tokenization(request)
def tokenization(request: Request) -> OpenAIServingTokenization:
return request.app.state.openai_serving_tokenization
def tokenization(request: Request) -> ServingTokenization:
return request.app.state.serving_tokenization
def engine_client(request: Request) -> EngineClient: