DeepSpeed
Enabling Muon Optimizer in DeepSpeed
#7509
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
32
Changes
View On
GitHub
Enabling Muon Optimizer in DeepSpeed
#7509
PKUWZP
merged 32 commits into
deepspeedai:master
from
pengdurice:peng-add-muon-v1
add code changes and push to my branch
0d2ac8bb
bring back some comments
b43a4757
fix default use_muon to be backward compatible
8fd098ff
fix some issues at copying code from test to branch
43c6db8c
add muon change
1098b598
add unit test case
abd43284
change wording
bc42a55f
assert optimizer
d80613b5
make sure initialization works
98c7f969
enable only updating every LGA steps
68941d95
enable only updating every LGA steps
3e753722
Merge branch 'deepspeedai:master' into peng-add-muon-v1
f33e7c70
PKUWZP
requested a review
from
sfc-gh-truwase
200 days ago
PKUWZP
requested a review
from
tjruwase
200 days ago
PKUWZP
requested a review
from
tohtana
200 days ago
PKUWZP
requested a review
from
loadams
200 days ago
Clarify Muon dependency in setup.py
c894dcc9
try to add to install_requires and see if it fix
d5fd9254
fix install requires and add copyright for
aa408457
Fix formatting in setup.py for muon dependency
e081773f
fix conflicts
9c5e344d
use original muon directly in the code
16a9d73c
use original muon directly in the code, fix deepspeed comm error and …
9d552181
break the torch distributed error
d9988018
add licence
057c9f25
add licence
64a02324
add licence fix yapf
8ba1f063
Fix the end-of-file error.
bfd92605
Fix the end-of-file formatting error.
044d9e82
afix eof error
35ef8e3d
Fix the License and Copyright.
720e2d4d
Fix the MIT License for original Muon Implementation.
e695af93
Fix the license issue in original Muon Implementation.
08b5c08d
Fix Copyright in test_muon.py
363cbc68
Merge branch 'master' into peng-add-muon-v1
a8881260
sfc-gh-truwase
commented on 2025-08-25
sfc-gh-truwase
commented on 2025-08-25
sfc-gh-truwase
approved these changes on 2025-08-26
Merge branch 'master' into peng-add-muon-v1
a2c95b0b
PKUWZP
merged
66ad2780
into master
197 days ago
Login to write a write a comment.
Login via GitHub
Reviewers
sfc-gh-truwase
pengdurice
tjruwase
tohtana
loadams
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Login to write a write a comment.
Login via GitHub