[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
@@ -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