xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-04-08 13:07:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix pre-commit

Browse Source 
Signed-off-by: Robert Shaw <rshaw@neuralmagic.com>
This commit is contained in:
Robert Shaw 2025-03-23 09:38:55 -04:00
parent 2ba687d39f
commit 79e465f557
1 changed files with 19 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
vllm/entrypoints/disaggregated/types.py
View File

@ -1,7 +1,9 @@
# SPDX-License-Identifier: Apache-2.0
from typing import Optional
import msgspec
from typing import List, Optional
from vllm import SamplingParams
# NOTE FOR DEVELOPERS:
@ -9,23 +11,28 @@ from vllm import SamplingParams
# SETUP WE WILL USE TCP. WE CANNOT USE PICKLE OTHERWISE
# WE RISK REMOTE CODE EXECUTION FROM UNSTRUSTED USERS.
class PDRequest(msgspec.Struct,
array_like=True, # type: ignore[call-arg]
omit_defaults=True, # type: ignore[call-arg]
gc=False): # type: ignore[call-arg]
class PDRequest(
msgspec.Struct,
array_like=True, # type: ignore[call-arg]
omit_defaults=True, # type: ignore[call-arg]
gc=False): # type: ignore[call-arg]
request_id: str
prompt_token_ids: List[int]
prompt_token_ids: list[int]
sampling_params: SamplingParams
# TODO: support multimodal inputs.
class PDResponse(msgspec.Struct,
array_like=True, # type: ignore[call-arg]
omit_defaults=True, # type: ignore[call-arg]
gc=False): # type: ignore[call-arg]
class PDResponse(
msgspec.Struct,
array_like=True, # type: ignore[call-arg]
omit_defaults=True, # type: ignore[call-arg]
gc=False): # type: ignore[call-arg]
request_id: str
success: bool
text: str
token_ids: List[int]
token_ids: list[int]
finish_reason: Optional[str] = None
stop_reason: Optional[str] = None
logprobs = None # TODO
# TODO: support full protocol.
logprobs = None