[Frontend] Refactor ServingTokenization entrypoint. (#46022)
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
This commit is contained in:
@@ -22,8 +22,9 @@ from vllm.entrypoints.openai.engine.protocol import (
|
||||
RequestResponseMetadata,
|
||||
UsageInfo,
|
||||
)
|
||||
from vllm.entrypoints.openai.engine.serving import OpenAIServing, SpeechToTextRequest
|
||||
from vllm.entrypoints.openai.engine.serving import OpenAIServing
|
||||
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
|
||||
from vllm.entrypoints.serve.engine.typing import SpeechToTextRequest
|
||||
from vllm.entrypoints.serve.utils.api_utils import get_max_tokens
|
||||
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
|
||||
from vllm.exceptions import VLLMValidationError
|
||||
|
||||
Reference in New Issue
Block a user