[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
@@ -15,12 +15,12 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
)
from vllm.entrypoints.openai.chat_completion.serving import OpenAIServingChat
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.orca_metrics import metrics_header
from vllm.entrypoints.openai.utils import validate_json_request
from vllm.entrypoints.utils import (
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
with_cancellation,
)
from vllm.entrypoints.serve.utils.orca_metrics import metrics_header
from vllm.logger import init_logger
logger = init_logger(__name__)
@@ -21,7 +21,7 @@ from vllm.entrypoints.openai.engine.protocol import (
RequestResponseMetadata,
UsageInfo,
)
from vllm.entrypoints.utils import get_max_tokens
from vllm.entrypoints.serve.utils.api_utils import get_max_tokens
from vllm.inputs import EngineInput
from vllm.logger import init_logger
from vllm.outputs import RequestOutput
@@ -21,7 +21,6 @@ from vllm.entrypoints.chat_utils import (
get_tool_call_id_type,
make_tool_call_id,
)
from vllm.entrypoints.logger import RequestLogger
from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionLogProb,
ChatCompletionLogProbs,
@@ -57,8 +56,11 @@ from vllm.entrypoints.openai.parser.harmony_utils import (
get_streamable_parser_for_assistant,
parse_chat_output,
)
from vllm.entrypoints.openai.utils import maybe_filter_parallel_tool_calls
from vllm.entrypoints.utils import get_max_tokens, should_include_usage
from vllm.entrypoints.serve.utils.api_utils import get_max_tokens, should_include_usage
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.entrypoints.serve.utils.tool_calls_utils import (
maybe_filter_parallel_tool_calls,
)
from vllm.inputs import EngineInput
from vllm.logger import init_logger
from vllm.logprobs import Logprob