A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[BugFix] Fix Python 3.9 Support (#23306)
Signed-off-by: Jared O'Connell <46976761+jaredoconnell@users.noreply.github.com> Signed-off-by: Cyrus Leung <cyrus.tl.leung@gmail.com> Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
Jared O'Connell committed
Aug 21, 2025 at 02:23 UTC
31282401b6c5c4271eb558e6af3f1d42b1d157c8
1 file changed
+2
-2
vllm/benchmarks/lib/endpoint_request_func.py
+2
-2
@@ -9,7 +9,7 @@ import sys
9
import time
10
import traceback
11
from dataclasses import dataclass, field
12
-from typing import Optional
12
+from typing import Optional, Union
13
14
import aiohttp
15
from tqdm.asyncio import tqdm
@@ -28,7 +28,7 @@ class RequestFuncInput:
28
model_name: Optional[str] = None
29
logprobs: Optional[int] = None
30
extra_body: Optional[dict] = None
31
- multi_modal_content: Optional[dict | list[dict]] = None
31
+ multi_modal_content: Optional[Union[dict, list[dict]]] = None
32
ignore_eos: bool = False
33
language: Optional[str] = None
34
request_id: Optional[str] = None