Skip to content

[Deepspeed] skip device mesh creation when deepspeed and sp_size >1 #3914

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
SunMarc merged 1 commit into from
Jan 27, 2026
Merged

[Deepspeed] skip device mesh creation when deepspeed and sp_size >1 #3914

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions src/accelerate/parallelism_config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -215,6 +215,11 @@ def build_device_mesh(self, device_type: str):
Args:
device_type (`str`): The type of device for which to build the mesh, e
"""
# Skip mesh creation for DeepSpeed SP - DeepSpeed handles its own SP groups
# Only skip when SP is actually enabled (sp_size > 1), otherwise user might still want TP/CP/FSDP
if self.sp_backend == "deepspeed" and self.sp_size > 1:
return None

if is_torch_version(">=", "2.2.0"):
from torch.distributed.device_mesh import init_device_mesh
else:
Expand Down
Loading