frozenleaves/vllm
1
0
Fork 0
mirror of https://github.com/vllm-project/vllm.git synced 2025-10-20 14:53:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Set VLLM_ALLREDUCE_USE_SYMM_MEM default to False (#24696)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye
2025-09-11 17:32:27 -04:00
committed by GitHub
parent e42af78b18
commit 1ec20355f5
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/envs.py
View File

@ -1218,7 +1218,7 @@ environment_variables: dict[str, Callable[[], Any]] = {
# Whether to use pytorch symmetric memory for allreduce
"VLLM_ALLREDUCE_USE_SYMM_MEM":
lambda: bool(int(os.getenv("VLLM_ALLREDUCE_USE_SYMM_MEM", "1"))),
lambda: bool(int(os.getenv("VLLM_ALLREDUCE_USE_SYMM_MEM", "0"))),
# Allows vllm to find tuned config under customized folder
"VLLM_TUNED_CONFIG_FOLDER":