20231088/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

[Bugfix] Fix default behavior/fallback for pp in v1 (#16057)

Browse Source 
Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin 2025-04-04 11:58:08 -06:00 committed by GitHub
parent a6d042df0a
commit 4708f13a9c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
vllm/engine/arg_utils.py
View File

@ -1521,8 +1521,9 @@ class EngineArgs:
# PP is supported on V1 with Ray distributed executor, # PP is supported on V1 with Ray distributed executor,
# but off for MP distributed executor for now. # but off for MP distributed executor for now.
if (self.pipeline_parallel_size > 1 if (self.pipeline_parallel_size > 1
and self.distributed_executor_backend == "mp" and self.distributed_executor_backend != "ray"):
and _warn_or_fallback("PP (MP distributed executor)")): name = "Pipeline Parallelism without Ray distributed executor"
_raise_or_fallback(feature_name=name, recommend_to_remove=False)
return False return False
# ngram is supported on V1, but off by default for now. # ngram is supported on V1, but off by default for now.