[Render] Add /derender endpoints for disaggregated postprocessing (#43606)

Signed-off-by: Martin Hickey <martin.hickey@ie.ibm.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Martin Hickey
2026-06-13 13:55:33 +08:00
committed by GitHub
co-authored by Isotr0py
parent 1033ffac2e
commit 43f0e024bc
7 changed files with 898 additions and 9 deletions
+61 -3
View File
@@ -5,10 +5,20 @@ from http import HTTPStatus
from fastapi import APIRouter, Depends, FastAPI, Request
from fastapi.responses import JSONResponse
from vllm.entrypoints.openai.chat_completion.protocol import ChatCompletionRequest
from vllm.entrypoints.openai.completion.protocol import CompletionRequest
from vllm.entrypoints.openai.chat_completion.protocol import (
ChatCompletionRequest,
ChatCompletionResponse,
)
from vllm.entrypoints.openai.completion.protocol import (
CompletionRequest,
CompletionResponse,
)
from vllm.entrypoints.openai.engine.protocol import ErrorResponse
from vllm.entrypoints.serve.disagg.protocol import GenerateRequest
from vllm.entrypoints.serve.disagg.protocol import (
DerenderChatRequest,
DerenderCompletionRequest,
GenerateRequest,
)
from vllm.entrypoints.serve.render.serving import OpenAIServingRender
from vllm.entrypoints.serve.utils.api_utils import validate_json_request
from vllm.logger import init_logger
@@ -71,5 +81,53 @@ async def render_completion(request: CompletionRequest, raw_request: Request):
return JSONResponse(content=[item.model_dump() for item in result])
@router.post(
"/v1/chat/completions/derender",
dependencies=[Depends(validate_json_request)],
response_model=ChatCompletionResponse,
responses={
HTTPStatus.BAD_REQUEST.value: {"model": ErrorResponse},
HTTPStatus.NOT_FOUND.value: {"model": ErrorResponse},
HTTPStatus.INTERNAL_SERVER_ERROR.value: {"model": ErrorResponse},
},
)
async def derender_chat_completion(request: DerenderChatRequest, raw_request: Request):
handler = render(raw_request)
if handler is None:
raise NotImplementedError(
"The model does not support Chat Completions Derender API"
)
result = await handler.derender_chat_response(request)
if isinstance(result, ErrorResponse):
return JSONResponse(content=result.model_dump(), status_code=result.error.code)
return JSONResponse(content=result.model_dump())
@router.post(
"/v1/completions/derender",
dependencies=[Depends(validate_json_request)],
response_model=CompletionResponse,
responses={
HTTPStatus.BAD_REQUEST.value: {"model": ErrorResponse},
HTTPStatus.NOT_FOUND.value: {"model": ErrorResponse},
HTTPStatus.INTERNAL_SERVER_ERROR.value: {"model": ErrorResponse},
},
)
async def derender_completion(request: DerenderCompletionRequest, raw_request: Request):
handler = render(raw_request)
if handler is None:
raise NotImplementedError("The model does not support Completions Derender API")
result = await handler.derender_completion_response(request)
if isinstance(result, ErrorResponse):
return JSONResponse(content=result.model_dump(), status_code=result.error.code)
return JSONResponse(content=result.model_dump())
def attach_router(app: FastAPI) -> None:
app.include_router(router)