xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-06-01 05:27:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

temporarily remove enable_microbatching

Browse Source 
Signed-off-by: Sage Moore <sage@neuralmagic.com>
This commit is contained in:
Sage Moore 2025-06-02 19:30:01 +00:00
parent 919eef995b
commit 2731e8cbcb
1 changed files with 0 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
vllm/config.py
View File

@ -1741,15 +1741,6 @@ class ParallelConfig:
rank: int = 0 rank: int = 0
"""Global rank in distributed setup.""" """Global rank in distributed setup."""
always_microbatch_if_enabled: bool = True
"""Always microbatch if microbatching is enabled. Easier to sync between
dp workers."""
microbatching_token_threshold: int = 4
"""The threshold for microbatching. If the number of tokens in the
request is greater than this threshold, microbatching will be used.
Otherwise, the request will be processed in a single batch."""
@property @property
def world_size_across_dp(self) -> int: def world_size_across_dp(self) -> int:
"""world_size_across_dp is TPxPPxDP, it is the size of the world """world_size_across_dp is TPxPPxDP, it is the size of the world