Skip to content

Commit

Permalink
[Serve] Fix the multiplex doc format (ray-project#39497)
Browse files Browse the repository at this point in the history
  • Loading branch information
sihanwang41 authored Sep 8, 2023
1 parent e6abe26 commit b66e56e
Showing 1 changed file with 5 additions and 5 deletions.
10 changes: 5 additions & 5 deletions python/ray/serve/api.py
Original file line number Diff line number Diff line change
Expand Up @@ -698,11 +698,11 @@ async def __call__(self, request):
Args:
max_num_models_per_replica: the maximum number of models
to be loaded on each replica. By default, it is 3, which
means that each replica can cache up to 3 models. You can
set it to a larger number if you have enough memory on
the node resource, in opposite, you can set it to a smaller
number if you want to save memory on the node resource.
to be loaded on each replica. By default, it is 3, which
means that each replica can cache up to 3 models. You can
set it to a larger number if you have enough memory on
the node resource, in opposite, you can set it to a smaller
number if you want to save memory on the node resource.
"""

if func is not None:
Expand Down

0 comments on commit b66e56e

Please sign in to comment.