[Frontend] Move engine/protocol.py out openai folder (#54492)

Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Signed-off-by: wang.yuqi <noooop@126.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
This commit is contained in:
wang.yuqi
2026-08-31 10:13:50 +00:00
committed by GitHub
co-authored by Cyrus Leung
parent d8de4ae322
commit eeb549a74d
194 changed files with 550 additions and 585 deletions
+1 -1
View File
@@ -16,7 +16,7 @@ from vllm.entrypoints.anthropic.protocol import (
AnthropicMessagesResponse,
)
from vllm.entrypoints.anthropic.serving import AnthropicServingMessages
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.exception_handling.error_response import (
create_error_response,
)
+6 -7
View File
@@ -30,6 +30,11 @@ from vllm.entrypoints.anthropic.protocol import (
AnthropicUsage,
)
from vllm.entrypoints.chat_utils import ChatTemplateContentFormatOption
from vllm.entrypoints.generate.base.protocol import (
JsonSchemaResponseFormat,
ResponseFormat,
StreamOptions,
)
from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionNamedToolChoiceParam,
ChatCompletionRequest,
@@ -38,14 +43,8 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionToolsParam,
)
from vllm.entrypoints.openai.chat_completion.serving import OpenAIServingChat
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
JsonSchemaResponseFormat,
ResponseFormat,
StreamOptions,
UsageInfo,
)
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import ErrorResponse, UsageInfo
from vllm.entrypoints.serve.exception_handling.utils import sanitize_message
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.renderers.online_renderer import OnlineRenderer
+1 -1
View File
@@ -38,7 +38,7 @@ from fastapi.responses import JSONResponse, Response, StreamingResponse
from starlette.middleware.base import BaseHTTPMiddleware
import vllm.envs as envs
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.exception_handling.utils import sanitize_message
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
@@ -29,11 +29,9 @@ from typing import Any, Literal
from pydantic import Field, model_serializer
from vllm.entrypoints.generate.base.protocol import DeltaMessage
from vllm.entrypoints.openai.chat_completion.protocol import ChatMessage
from vllm.entrypoints.openai.engine.protocol import (
DeltaMessage,
OpenAIBaseModel,
)
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel
class CitationSource(OpenAIBaseModel):
+6 -7
View File
@@ -71,6 +71,11 @@ from vllm.entrypoints.cohere.protocol import (
ToolCallStartEvent,
ToolPlanDeltaEvent,
)
from vllm.entrypoints.generate.base.protocol import (
JsonSchemaResponseFormat,
ResponseFormat,
StreamOptions,
)
from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionRequest,
ChatCompletionResponse,
@@ -79,13 +84,8 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
ChatMessage,
)
from vllm.entrypoints.openai.chat_completion.serving import OpenAIServingChat
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
JsonSchemaResponseFormat,
ResponseFormat,
StreamOptions,
)
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import ErrorInfo, ErrorResponse
from vllm.entrypoints.serve.exception_handling.utils import sanitize_message
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.parser.abstract_parser import Parser
@@ -1761,7 +1761,6 @@ class CohereServingChatV2(OpenAIServingChat):
# so the router can translate the envelope uniformly. ``param``
# is accepted for signature parity with the base class but is
# not surfaced in the Cohere wire format.
from vllm.entrypoints.openai.engine.protocol import ErrorInfo
del param # unused; kept for signature compatibility
return ErrorResponse(
@@ -4,9 +4,7 @@
# Adapted from
# https://github.com/lm-sys/FastChat/blob/168ccc29d3f7edc50823016105c024fe2282732a/fastchat/protocol/openai_api_protocol.py
import json
import time
from http import HTTPStatus
from typing import Annotated, Any, ClassVar, Literal, TypeAlias
from typing import Annotated, Any, Literal, TypeAlias
import regex as re
from pydantic import (
@@ -14,16 +12,15 @@ from pydantic import (
ConfigDict,
Field,
model_serializer,
model_validator,
)
import vllm.envs as envs
from vllm.config.utils import replace
from vllm.entrypoints.chat_utils import make_tool_call_id
from vllm.exceptions import VLLMServerError, VLLMValidationError
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
from vllm.sampling_params import StructuredOutputsParams
from vllm.utils import random_uuid
from vllm.utils.import_utils import resolve_obj_by_qualname
logger = init_logger(__name__)
@@ -33,102 +30,6 @@ StopParam: TypeAlias = (
)
class OpenAIBaseModel(BaseModel):
# OpenAI API does allow extra fields
model_config = ConfigDict(extra="allow")
# Cache class field names
field_names: ClassVar[set[str] | None] = None
@model_validator(mode="wrap")
@classmethod
def __log_extra_fields__(cls, data, handler):
result = handler(data)
if not isinstance(data, dict):
return result
field_names = cls.field_names
if field_names is None:
# Get all class field names and their potential aliases
field_names = set()
for field_name, field in cls.model_fields.items():
field_names.add(field_name)
if alias := getattr(field, "alias", None):
field_names.add(alias)
cls.field_names = field_names
# Compare against both field names and aliases
if any(k not in field_names for k in data):
logger.debug(
"The following fields were present in the request but ignored: %s",
data.keys() - field_names,
)
return result
class ErrorInfo(OpenAIBaseModel):
message: str
type: str
param: str | None = None
code: int
class ErrorResponse(OpenAIBaseModel):
error: ErrorInfo
class ModelPermission(OpenAIBaseModel):
id: str = Field(default_factory=lambda: f"modelperm-{random_uuid()}")
object: str = "model_permission"
created: int = Field(default_factory=lambda: int(time.time()))
allow_create_engine: bool = False
allow_sampling: bool = True
allow_logprobs: bool = True
allow_search_indices: bool = False
allow_view: bool = True
allow_fine_tuning: bool = False
organization: str = "*"
group: str | None = None
is_blocking: bool = False
class ModelCard(OpenAIBaseModel):
id: str
object: str = "model"
created: int = Field(default_factory=lambda: int(time.time()))
owned_by: str = "vllm"
root: str | None = None
parent: str | None = None
max_model_len: int | None = None
permission: list[ModelPermission] = Field(default_factory=list)
class ModelList(OpenAIBaseModel):
object: str = "list"
data: list[ModelCard] = Field(default_factory=list)
class PromptTokenUsageInfo(OpenAIBaseModel):
cached_tokens: int | None = None
created_cache_tokens: int | None = None
multimodal_tokens: dict[str, int] | None = None
"""Prompt tokens contributed by each input modality, keyed by modality name
(e.g. `image`, `audio`, `video`). A breakdown of the multimodal
placeholder tokens already counted in `prompt_tokens`; `None` when the
request has no multimodal input."""
class CompletionTokenUsageInfo(OpenAIBaseModel):
reasoning_tokens: int = 0
class UsageInfo(OpenAIBaseModel):
prompt_tokens: int = 0
total_tokens: int = 0
completion_tokens: int | None = 0
prompt_tokens_details: PromptTokenUsageInfo | None = None
completion_tokens_details: CompletionTokenUsageInfo | None = None
class SpeculativeDecodingMetrics(OpenAIBaseModel):
"""Per-request speculative-decoding acceptance metrics.
@@ -437,11 +338,3 @@ class DeltaMessage(OpenAIBaseModel):
if len(data.get("tool_calls", [])) == 0:
data.pop("tool_calls", None)
return data
class GenerationError(VLLMServerError):
"""raised when finish_reason indicates internal server error (500)"""
def __init__(self, message: str = "Internal server error"):
super().__init__(message)
self.status_code = HTTPStatus.INTERNAL_SERVER_ERROR
+8 -10
View File
@@ -5,27 +5,28 @@ import time
from collections.abc import Awaitable, Mapping
from dataclasses import dataclass, field
from http import HTTPStatus
from typing import TYPE_CHECKING, ClassVar, Generic, TypeVar
from typing import ClassVar, Generic, TypeVar
from fastapi import Request
from pydantic import ConfigDict
from starlette.datastructures import Headers
from vllm import RequestOutput
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.generate.beam_search.online import BeamSearchOnlineMixin
from vllm.entrypoints.openai.chat_completion.protocol import ChatCompletionRequest
from vllm.entrypoints.openai.completion.protocol import CompletionRequest
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
GenerationError,
from vllm.entrypoints.generate.base.protocol import (
PerRequestMetrics,
SpeculativeDecodingMetrics,
)
from vllm.entrypoints.generate.beam_search.online import BeamSearchOnlineMixin
from vllm.entrypoints.openai.chat_completion.protocol import ChatCompletionRequest
from vllm.entrypoints.openai.completion.protocol import CompletionRequest
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.openai.responses.protocol import ResponsesRequest
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.engine.typing import AnyRequest
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.exceptions import GenerationError
from vllm.inputs import EngineInput
from vllm.logger import init_logger
from vllm.logprobs import Logprob, PromptLogprobs
@@ -38,9 +39,6 @@ from vllm.tracing import (
)
from vllm.v1.metrics.stats import RequestStateStats
if TYPE_CHECKING:
from vllm.outputs import RequestOutput
logger = init_logger(__name__)
RequestT = TypeVar("RequestT", bound=AnyRequest)
@@ -9,7 +9,7 @@ from vllm.entrypoints.generate.generative_scoring.serving import (
GenerativeScoringResponse,
ServingGenerativeScoring,
)
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
@@ -18,12 +18,12 @@ from fastapi import Request
from pydantic import Field
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import (
ErrorResponse,
OpenAIBaseModel,
UsageInfo,
)
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.inputs import EngineInput, tokens_input
+5 -5
View File
@@ -35,11 +35,6 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionRequest,
ChatCompletionResponse,
)
from vllm.entrypoints.openai.engine.protocol import (
ErrorInfo,
ErrorResponse,
OpenAIBaseModel,
)
from vllm.entrypoints.pooling.embed.protocol import (
EmbeddingRequest,
EmbeddingResponse,
@@ -51,6 +46,11 @@ from vllm.entrypoints.pooling.scoring.protocol import (
ScoreResponse,
)
from vllm.entrypoints.serve import create_error_response
from vllm.entrypoints.serve.engine.protocol import (
ErrorInfo,
ErrorResponse,
OpenAIBaseModel,
)
from vllm.entrypoints.speech_to_text.transcription.protocol import (
TranscriptionRequest,
TranscriptionResponse,
@@ -14,8 +14,8 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionResponse,
)
from vllm.entrypoints.openai.chat_completion.serving import OpenAIServingChat
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.sse_keep_alive import with_sse_keep_alive
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
@@ -9,6 +9,7 @@ from http import HTTPStatus
from fastapi import Request
from vllm.entrypoints.chat_utils import ConversationMessage
from vllm.entrypoints.generate.base.protocol import RequestResponseMetadata
from vllm.entrypoints.openai.chat_completion.protocol import (
BatchChatCompletionRequest,
ChatCompletionResponse,
@@ -16,11 +17,7 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
ChatMessage,
)
from vllm.entrypoints.openai.chat_completion.serving import OpenAIServingChat
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
RequestResponseMetadata,
UsageInfo,
)
from vllm.entrypoints.serve.engine.protocol import ErrorResponse, UsageInfo
from vllm.entrypoints.serve.utils.api_utils import get_max_tokens
from vllm.inputs import EngineInput
from vllm.logger import init_logger
@@ -23,21 +23,20 @@ from vllm.entrypoints.chat_utils import (
ChatCompletionMessageParam,
ChatTemplateContentFormatOption,
)
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.generate.base.protocol import (
AnyResponseFormat,
DeltaMessage,
FunctionCall,
FunctionDefinition,
OpenAIBaseModel,
PerRequestMetrics,
StopParam,
StreamOptions,
ToolCall,
UsageInfo,
structured_outputs_from_response_format,
validate_structural_tag_response_format,
validate_structured_outputs_structural_tag,
)
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
from vllm.logprobs import Logprob
@@ -16,9 +16,15 @@ from vllm.entrypoints.chat_utils import (
ConversationMessage,
make_tool_call_id,
)
from vllm.entrypoints.generate.base.protocol import (
DeltaMessage,
FunctionCall,
PerRequestMetrics,
RequestResponseMetadata,
ToolCall,
)
from vllm.entrypoints.generate.base.serving import (
GenerateBaseServing,
GenerationError,
build_per_request_timing_metrics,
build_spec_decoding_metrics,
clamp_prompt_logprobs,
@@ -36,23 +42,19 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionStreamResponse,
ChatMessage,
)
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import (
CompletionTokenUsageInfo,
DeltaMessage,
ErrorResponse,
FunctionCall,
PerRequestMetrics,
PromptTokenUsageInfo,
RequestResponseMetadata,
ToolCall,
UsageInfo,
)
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
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.exceptions import GenerationError
from vllm.inputs import EngineInput, MultiModalPlaceholders
from vllm.logger import init_logger
from vllm.logprobs import Logprob
@@ -12,8 +12,8 @@ from vllm.entrypoints.openai.completion.protocol import (
CompletionResponse,
)
from vllm.entrypoints.openai.completion.serving import OpenAIServingCompletion
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.sse_keep_alive import with_sse_keep_alive
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
@@ -10,17 +10,16 @@ from pydantic import Field, model_validator
import vllm.envs as envs
from vllm.config import ModelConfig
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.generate.base.protocol import (
AnyResponseFormat,
OpenAIBaseModel,
PerRequestMetrics,
StopParam,
StreamOptions,
UsageInfo,
structured_outputs_from_response_format,
validate_structural_tag_response_format,
validate_structured_outputs_structural_tag,
)
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
from vllm.logprobs import Logprob
@@ -10,9 +10,12 @@ from typing import cast
from fastapi import Request
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.generate.base.protocol import (
PerRequestMetrics,
RequestResponseMetadata,
)
from vllm.entrypoints.generate.base.serving import (
GenerateBaseServing,
GenerationError,
build_per_request_timing_metrics,
build_spec_decoding_metrics,
clamp_prompt_logprobs,
@@ -26,17 +29,15 @@ from vllm.entrypoints.openai.completion.protocol import (
CompletionResponseStreamChoice,
CompletionStreamResponse,
)
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import (
ErrorResponse,
PerRequestMetrics,
PromptTokenUsageInfo,
RequestResponseMetadata,
UsageInfo,
)
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
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.exceptions import VLLMValidationError
from vllm.exceptions import GenerationError, VLLMValidationError
from vllm.inputs import EngineInput
from vllm.logger import init_logger
from vllm.logprobs import Logprob
@@ -1,2 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+3 -3
View File
@@ -7,14 +7,14 @@ from http import HTTPStatus
from vllm.config import ModelConfig
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.openai.models.protocol import BaseModelPath, LoRAModulePath
from vllm.entrypoints.serve import create_error_response
from vllm.entrypoints.serve.engine.protocol import (
ErrorResponse,
ModelCard,
ModelList,
ModelPermission,
)
from vllm.entrypoints.openai.models.protocol import BaseModelPath, LoRAModulePath
from vllm.entrypoints.serve import create_error_response
from vllm.entrypoints.serve.lora.protocol import (
LoadLoRAAdapterRequest,
UnloadLoRAAdapterRequest,
@@ -8,13 +8,13 @@ from http import HTTPStatus
from fastapi import APIRouter, Depends, FastAPI, Request
from fastapi.responses import JSONResponse, StreamingResponse
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.responses.protocol import (
ResponsesRequest,
ResponsesResponse,
StreamingResponsesResponse,
)
from vllm.entrypoints.openai.responses.serving import OpenAIServingResponses
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
+1 -3
View File
@@ -24,11 +24,9 @@ from vllm import envs
from vllm.entrypoints.chat_utils import (
ChatTemplateContentFormatOption,
)
from vllm.entrypoints.generate.base.protocol import FunctionCall
from vllm.entrypoints.mcp.tool import Tool
from vllm.entrypoints.mcp.tool_server import ToolServer
from vllm.entrypoints.openai.engine.protocol import (
FunctionCall,
)
from vllm.entrypoints.openai.parser.harmony_utils import render_for_completion
from vllm.entrypoints.openai.responses.protocol import (
ResponseInputOutputItem,
@@ -60,7 +60,8 @@ from vllm.entrypoints.chat_utils import (
ChatCompletionMessageParam,
ChatTemplateContentFormatOption,
)
from vllm.entrypoints.openai.engine.protocol import OpenAIBaseModel, StopParam
from vllm.entrypoints.generate.base.protocol import StopParam
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
from vllm.renderers import ChatParams, TokenizeParams, merge_kwargs
+5 -8
View File
@@ -31,16 +31,12 @@ from vllm.entrypoints.chat_utils import (
ChatCompletionMessageParam,
ChatTemplateContentFormatOption,
)
from vllm.entrypoints.generate.base.serving import (
GenerateBaseServing,
GenerationError,
)
from vllm.entrypoints.mcp.tool_server import ToolServer
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.generate.base.protocol import (
DeltaMessage,
ErrorResponse,
RequestResponseMetadata,
)
from vllm.entrypoints.generate.base.serving import GenerateBaseServing
from vllm.entrypoints.mcp.tool_server import ToolServer
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.openai.parser.harmony_utils import (
build_harmony_preamble,
@@ -89,9 +85,10 @@ from vllm.entrypoints.openai.responses.utils import (
extract_function_tool_names,
extract_tool_types,
)
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
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
from vllm.exceptions import GenerationError, VLLMValidationError
from vllm.inputs import EngineInput, tokens_input
from vllm.logger import init_logger
from vllm.logprobs import Logprob as SampleLogprob
@@ -61,8 +61,8 @@ from openai.types.responses.response_reasoning_item import (
from openai.types.responses.tool import Tool
from openai_harmony import Message as HarmonyMessage
from vllm.entrypoints.generate.base.protocol import DeltaMessage, DeltaToolCall
from vllm.entrypoints.mcp.tool_server import ToolServer
from vllm.entrypoints.openai.engine.protocol import DeltaMessage, DeltaToolCall
from vllm.entrypoints.openai.parser.harmony_utils import (
extract_function_from_recipient,
is_function_recipient,
+1 -1
View File
@@ -31,11 +31,11 @@ from openai.types.responses.tool import Tool
from vllm import envs
from vllm.entrypoints.chat_utils import make_tool_call_id
from vllm.entrypoints.generate.base.protocol import FunctionCall, FunctionDefinition
from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionMessageParam,
ChatCompletionToolsParam,
)
from vllm.entrypoints.openai.engine.protocol import FunctionCall, FunctionDefinition
from vllm.entrypoints.openai.responses.protocol import ResponseInputOutputItem
from vllm.logger import init_logger
from vllm.tool_parsers.utils import (
+1 -1
View File
@@ -11,7 +11,7 @@ from vllm.entrypoints.chat_utils import (
ChatCompletionMessageParam,
ChatTemplateContentFormatOption,
)
from vllm.entrypoints.openai.engine.protocol import OpenAIBaseModel
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel
from vllm.exceptions import VLLMValidationError
from vllm.renderers import ChatParams, TokenizeParams, merge_kwargs
from vllm.tasks import check_removed_pooling_task
+1 -1
View File
@@ -16,7 +16,6 @@ from vllm import PoolingRequestOutput, envs
from vllm.config import VllmConfig
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.chat_utils import ChatTemplateConfig
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.engine.typing import AnyRequest
@@ -30,6 +29,7 @@ from vllm.tracing import (
)
from vllm.utils.async_utils import make_async, merge_async_iterators
from ...serve.engine.protocol import ErrorResponse
from ..typing import AnyPoolingRequest, PoolingServeContext
from .io_processor import PoolingIOProcessor
@@ -7,7 +7,7 @@ from typing import Annotated, TypeAlias
from pydantic import BeforeValidator, Field
from vllm import PoolingParams
from vllm.entrypoints.openai.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.logger import init_logger
from vllm.utils import random_uuid
+1 -1
View File
@@ -6,7 +6,7 @@ from typing import TypeAlias
import numpy as np
from fastapi.responses import JSONResponse
from vllm.entrypoints.openai.engine.protocol import UsageInfo
from vllm.entrypoints.serve.engine.protocol import UsageInfo
from vllm.logger import init_logger
from vllm.outputs import ClassificationOutput
+1 -1
View File
@@ -5,7 +5,7 @@ from http import HTTPStatus
from fastapi import APIRouter, Depends, Request
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
+1 -1
View File
@@ -18,7 +18,7 @@ from pydantic import BaseModel, BeforeValidator, Field, model_validator
from vllm import PoolingParams
from vllm.entrypoints.chat_utils import ChatCompletionMessageParam
from vllm.entrypoints.openai.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.utils import random_uuid
from ..base.protocol import (
@@ -4,7 +4,7 @@ from http import HTTPStatus
from fastapi import APIRouter, Depends, Request
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
+1 -1
View File
@@ -7,7 +7,7 @@ from pydantic import BeforeValidator, Field
from vllm import PoolingParams
from vllm.config import ModelConfig
from vllm.entrypoints.openai.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.renderers import TokenizeParams
from vllm.tasks import PoolingTask
from vllm.utils import random_uuid
@@ -4,7 +4,7 @@ from http import HTTPStatus
from fastapi import APIRouter, Depends, Request
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
validate_json_request,
+1 -1
View File
@@ -7,7 +7,7 @@ from pydantic import BaseModel, Field, model_validator
from vllm import PoolingParams
from vllm.config import ModelConfig
from vllm.entrypoints.openai.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.renderers import TokenizeParams
from vllm.tasks import PoolingTask
from vllm.utils import random_uuid
+1 -1
View File
@@ -5,7 +5,7 @@ from fastapi.responses import JSONResponse, Response
from vllm import PoolingParams
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import UsageInfo
from vllm.entrypoints.serve.engine.protocol import UsageInfo
from vllm.logger import init_logger
from vllm.outputs import PoolingRequestOutput, ScoringRequestOutput
from vllm.tasks import SCORE_TYPE_MAP, SupportedTask
+1 -1
View File
@@ -14,7 +14,7 @@ import torch
from fastapi.responses import JSONResponse, StreamingResponse
from vllm.config import ModelConfig
from vllm.entrypoints.openai.engine.protocol import UsageInfo
from vllm.entrypoints.serve.engine.protocol import UsageInfo
from vllm.logger import init_logger
from vllm.outputs import PoolingRequestOutput
from vllm.tasks import SupportedTask
@@ -5,7 +5,7 @@ from http import HTTPStatus
from fastapi import APIRouter, Depends, Request
from fastapi.responses import JSONResponse
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import validate_json_request
from vllm.logger import init_logger
@@ -12,14 +12,11 @@ from vllm.entrypoints.openai.completion.protocol import (
CompletionResponse,
CompletionStreamResponse,
)
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
UsageInfo,
)
from vllm.entrypoints.openai.models.serving import (
OpenAIModelRegistry,
OpenAIServingModels,
)
from vllm.entrypoints.serve.engine.protocol import ErrorResponse, UsageInfo
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.inputs import (
@@ -8,7 +8,7 @@ from fastapi.responses import JSONResponse
from vllm.entrypoints.anthropic.protocol import AnthropicMessagesRequest
from vllm.entrypoints.openai.chat_completion.protocol import ChatCompletionRequest
from vllm.entrypoints.openai.completion.protocol import CompletionRequest
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import validate_json_request
from vllm.logger import init_logger
+1 -1
View File
@@ -6,7 +6,6 @@ from vllm.entrypoints.anthropic.protocol import AnthropicMessagesRequest
from vllm.entrypoints.anthropic.serving import AnthropicServingMessages
from vllm.entrypoints.openai.chat_completion.protocol import ChatCompletionRequest
from vllm.entrypoints.openai.completion.protocol import CompletionRequest
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.models.serving import (
OpenAIModelRegistry,
OpenAIServingModels,
@@ -17,6 +16,7 @@ from vllm.entrypoints.scale_out.token_in_token_out.protocol import (
MultiModalFeatures,
PlaceholderRangeInfo,
)
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.utils.api_utils import get_max_tokens
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
@@ -10,7 +10,6 @@ from fastapi import APIRouter, Depends, FastAPI, HTTPException, Request, Respons
from fastapi.responses import JSONResponse, StreamingResponse
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.tokenize.serving import ServingTokenization
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
@@ -19,6 +18,7 @@ from vllm.entrypoints.serve.utils.api_utils import (
)
from vllm.logger import init_logger
from ...serve.engine.protocol import ErrorResponse
from .protocol import (
GenerateRequest,
GenerateResponse,
@@ -11,6 +11,7 @@ from pydantic import (
)
from vllm.config import ModelConfig
from vllm.entrypoints.generate.base.protocol import StreamOptions
from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionLogProbs,
ChatCompletionRequest,
@@ -20,7 +21,7 @@ from vllm.entrypoints.openai.completion.protocol import (
CompletionRequest,
CompletionStreamResponse,
)
from vllm.entrypoints.openai.engine.protocol import StreamOptions, UsageInfo
from vllm.entrypoints.serve.engine.protocol import UsageInfo
from vllm.logprobs import Logprob
from vllm.renderers import TokenizeParams
from vllm.sampling_params import SamplingParams
@@ -12,6 +12,7 @@ from fastapi import Request
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.chat_utils import AsyncMultiModalItemTracker
from vllm.entrypoints.generate.base.protocol import RequestResponseMetadata
from vllm.entrypoints.generate.base.serving import (
GenerateBaseServing,
clamp_prompt_logprobs,
@@ -21,16 +22,15 @@ from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionLogProbs,
ChatCompletionLogProbsContent,
)
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import (
ErrorResponse,
GenerationError,
PromptTokenUsageInfo,
RequestResponseMetadata,
UsageInfo,
)
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
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.exceptions import GenerationError
from vllm.inputs import EngineInput, TokensPrompt, mm_input
from vllm.logger import init_logger
from vllm.logprobs import Logprob
@@ -9,10 +9,8 @@ from fastapi import APIRouter, Depends, FastAPI, HTTPException, Request
from fastapi.responses import JSONResponse
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
)
from vllm.entrypoints.serve.elastic_ep.middleware import get_scaling_elastic_ep
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import validate_json_request
from vllm.logger import init_logger
+115
View File
@@ -0,0 +1,115 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# Adapted from
# https://github.com/lm-sys/FastChat/blob/168ccc29d3f7edc50823016105c024fe2282732a/fastchat/protocol/openai_api_protocol.py
import time
from typing import ClassVar
from pydantic import (
BaseModel,
ConfigDict,
Field,
model_validator,
)
from vllm.logger import init_logger
from vllm.utils import random_uuid
logger = init_logger(__name__)
class OpenAIBaseModel(BaseModel):
# OpenAI API does allow extra fields
model_config = ConfigDict(extra="allow")
# Cache class field names
field_names: ClassVar[set[str] | None] = None
@model_validator(mode="wrap")
@classmethod
def __log_extra_fields__(cls, data, handler):
result = handler(data)
if not isinstance(data, dict):
return result
field_names = cls.field_names
if field_names is None:
# Get all class field names and their potential aliases
field_names = set()
for field_name, field in cls.model_fields.items():
field_names.add(field_name)
if alias := getattr(field, "alias", None):
field_names.add(alias)
cls.field_names = field_names
# Compare against both field names and aliases
if any(k not in field_names for k in data):
logger.debug(
"The following fields were present in the request but ignored: %s",
data.keys() - field_names,
)
return result
class ErrorInfo(OpenAIBaseModel):
message: str
type: str
param: str | None = None
code: int
class ErrorResponse(OpenAIBaseModel):
error: ErrorInfo
class ModelPermission(OpenAIBaseModel):
id: str = Field(default_factory=lambda: f"modelperm-{random_uuid()}")
object: str = "model_permission"
created: int = Field(default_factory=lambda: int(time.time()))
allow_create_engine: bool = False
allow_sampling: bool = True
allow_logprobs: bool = True
allow_search_indices: bool = False
allow_view: bool = True
allow_fine_tuning: bool = False
organization: str = "*"
group: str | None = None
is_blocking: bool = False
class ModelCard(OpenAIBaseModel):
id: str
object: str = "model"
created: int = Field(default_factory=lambda: int(time.time()))
owned_by: str = "vllm"
root: str | None = None
parent: str | None = None
max_model_len: int | None = None
permission: list[ModelPermission] = Field(default_factory=list)
class ModelList(OpenAIBaseModel):
object: str = "list"
data: list[ModelCard] = Field(default_factory=list)
class PromptTokenUsageInfo(OpenAIBaseModel):
cached_tokens: int | None = None
created_cache_tokens: int | None = None
multimodal_tokens: dict[str, int] | None = None
"""Prompt tokens contributed by each input modality, keyed by modality name
(e.g. `image`, `audio`, `video`). A breakdown of the multimodal
placeholder tokens already counted in `prompt_tokens`; `None` when the
request has no multimodal input."""
class CompletionTokenUsageInfo(OpenAIBaseModel):
reasoning_tokens: int = 0
class UsageInfo(OpenAIBaseModel):
prompt_tokens: int = 0
total_tokens: int = 0
completion_tokens: int | None = 0
prompt_tokens_details: PromptTokenUsageInfo | None = None
completion_tokens_details: CompletionTokenUsageInfo | None = None
+1 -1
View File
@@ -6,13 +6,13 @@ from fastapi import Request
from vllm import PromptType, SamplingParams, envs
from vllm.config import ModelConfig
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.models.serving import (
OpenAIModelRegistry,
OpenAIServingModels,
)
from vllm.entrypoints.pooling.typing import AnyPoolingRequest
from vllm.entrypoints.serve import create_error_response
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.typing import AnyRequest
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.exceptions import VLLMNotFoundError
@@ -2,13 +2,10 @@
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from http import HTTPStatus
from vllm.entrypoints.openai.engine.protocol import (
ErrorInfo,
ErrorResponse,
GenerationError,
)
from vllm.exceptions import GenerationError
from vllm.logger import init_logger
from ..engine.protocol import ErrorInfo, ErrorResponse
from .utils import sanitize_message
logger = init_logger(__name__)
@@ -5,7 +5,7 @@ from http import HTTPStatus
from fastapi import HTTPException, Request
from starlette.responses import JSONResponse
from vllm.entrypoints.openai.engine.protocol import ErrorInfo, ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorInfo, ErrorResponse
from vllm.logger import init_logger
from ..utils import sanitize_message
@@ -8,7 +8,7 @@ from fastapi import Request
from fastapi.exceptions import RequestValidationError
from starlette.responses import JSONResponse
from vllm.entrypoints.openai.engine.protocol import ErrorInfo, ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorInfo, ErrorResponse
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
@@ -5,8 +5,7 @@ from fastapi import Request
from starlette.responses import JSONResponse
from vllm.entrypoints.launchers.launcher import terminate_if_errored
from vllm.entrypoints.openai.engine.protocol import GenerationError
from vllm.exceptions import VLLMError
from vllm.exceptions import GenerationError, VLLMError
from vllm.logger import init_logger
from vllm.v1.engine.exceptions import EngineDeadError, EngineGenerateError
@@ -8,7 +8,7 @@ from fastapi import APIRouter, BackgroundTasks, Depends, FastAPI, HTTPException,
from fastapi.responses import JSONResponse
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import validate_json_request
from vllm.logger import init_logger
from vllm.v1.fault_tolerance.utils import FaultToleranceRequest
+1 -3
View File
@@ -7,11 +7,9 @@ from fastapi import APIRouter, Depends, FastAPI, Request
from fastapi.responses import JSONResponse, Response
from vllm import envs
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
)
from vllm.entrypoints.openai.models.api_router import models
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.lora.protocol import (
LoadLoRAAdapterRequest,
UnloadLoRAAdapterRequest,
@@ -12,8 +12,8 @@ from fastapi.responses import JSONResponse, Response
from vllm.config import ModelConfig
from vllm.entrypoints.generate.factories import get_generate_invocation_types
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.pooling.factories import get_pooling_invocation_types
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.instrumentator.basic import base
from vllm.entrypoints.serve.instrumentator.health import health
@@ -8,7 +8,7 @@ from fastapi import APIRouter, Depends, FastAPI, Request
from fastapi.responses import JSONResponse
from typing_extensions import assert_never
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.tokenize.protocol import (
DetokenizeRequest,
DetokenizeResponse,
+2 -6
View File
@@ -11,12 +11,8 @@ from vllm.entrypoints.chat_utils import (
ChatCompletionMessageParam,
ChatTemplateContentFormatOption,
)
from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionToolsParam,
)
from vllm.entrypoints.openai.engine.protocol import (
OpenAIBaseModel,
)
from vllm.entrypoints.openai.chat_completion.protocol import ChatCompletionToolsParam
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel
from vllm.exceptions import VLLMValidationError
from vllm.renderers import ChatParams, TokenizeParams, merge_kwargs
+1 -1
View File
@@ -6,11 +6,11 @@ from typing import Any, Final
from fastapi import Request
from vllm.entrypoints.chat_utils import ChatTemplateContentFormatOption
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.openai.models.serving import (
OpenAIModelRegistry,
OpenAIServingModels,
)
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.tokenize.protocol import (
DetokenizeRequest,
+1 -1
View File
@@ -17,7 +17,7 @@ from starlette.background import BackgroundTask, BackgroundTasks
from vllm import envs
from vllm.engine.arg_utils import EngineArgs
from vllm.entrypoints.openai.engine.protocol import StreamOptions
from vllm.entrypoints.generate.base.protocol import StreamOptions
from vllm.entrypoints.openai.models.protocol import LoRAModulePath
from vllm.logger import current_formatter_type, init_logger
from vllm.platforms import current_platform
@@ -16,14 +16,13 @@ from transformers import PreTrainedTokenizerBase
import vllm.envs as envs
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.generate.base.serving import GenerateBaseServing
from vllm.entrypoints.openai.engine.protocol import (
from vllm.entrypoints.generate.base.protocol import (
DeltaMessage,
ErrorResponse,
RequestResponseMetadata,
UsageInfo,
)
from vllm.entrypoints.generate.base.serving import GenerateBaseServing
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import ErrorResponse, UsageInfo
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
@@ -13,7 +13,7 @@ from fastapi import WebSocket
from starlette.websockets import WebSocketDisconnect
from vllm import envs
from vllm.entrypoints.openai.engine.protocol import ErrorResponse, UsageInfo
from vllm.entrypoints.serve.engine.protocol import ErrorResponse, UsageInfo
from vllm.entrypoints.serve.exception_handling.utils import sanitize_message
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
@@ -6,10 +6,7 @@ from typing import Literal
from pydantic import Field
from vllm.entrypoints.openai.engine.protocol import (
OpenAIBaseModel,
UsageInfo,
)
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.utils import random_uuid
# Client -> Server Events
@@ -8,7 +8,6 @@ from typing import Annotated
from fastapi import APIRouter, Form, Request
from fastapi.responses import JSONResponse, StreamingResponse
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
with_cancellation,
@@ -16,6 +15,7 @@ from vllm.entrypoints.serve.utils.api_utils import (
from vllm.entrypoints.speech_to_text.base.utils import read_upload_with_limit
from vllm.logger import init_logger
from ...serve.engine.protocol import ErrorResponse
from .protocol import TranscriptionRequest, TranscriptionResponseVariant
from .serving import OpenAIServingTranscription
@@ -13,11 +13,8 @@ from pydantic import (
)
from vllm.config.speech_to_text import SpeechToTextParams
from vllm.entrypoints.openai.engine.protocol import (
DeltaMessage,
OpenAIBaseModel,
UsageInfo,
)
from vllm.entrypoints.generate.base.protocol import DeltaMessage
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
from vllm.sampling_params import (
@@ -5,15 +5,13 @@ from collections.abc import AsyncGenerator
from fastapi import Request
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
RequestResponseMetadata,
)
from vllm.entrypoints.generate.base.protocol import RequestResponseMetadata
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.logger import init_logger
from vllm.outputs import RequestOutput
from ...serve.engine.protocol import ErrorResponse
from ..base.serving import SpeechToTextBaseServing
from .protocol import (
TranscriptionRequest,
@@ -8,7 +8,7 @@ from typing import Annotated
from fastapi import APIRouter, Form, Request
from fastapi.responses import JSONResponse, StreamingResponse
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.api_utils import (
load_aware_call,
with_cancellation,
@@ -12,11 +12,8 @@ from pydantic import (
)
from vllm.config.speech_to_text import SpeechToTextParams
from vllm.entrypoints.openai.engine.protocol import (
DeltaMessage,
OpenAIBaseModel,
UsageInfo,
)
from vllm.entrypoints.generate.base.protocol import DeltaMessage
from vllm.entrypoints.serve.engine.protocol import OpenAIBaseModel, UsageInfo
from vllm.exceptions import VLLMValidationError
from vllm.logger import init_logger
from vllm.sampling_params import (
@@ -5,11 +5,9 @@ from collections.abc import AsyncGenerator
from fastapi import Request
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.openai.engine.protocol import (
ErrorResponse,
RequestResponseMetadata,
)
from vllm.entrypoints.generate.base.protocol import RequestResponseMetadata
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.logger import init_logger
from vllm.outputs import RequestOutput