Skip to content

allow seperate learning rate "muon_lr" and "adam_lr" for muon optimizer #7658

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
merged 5 commits into from
Nov 11, 2025
Merged

allow seperate learning rate "muon_lr" and "adam_lr" for muon optimizer #7658

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
c94b822
allow seperate learning rate "muon_lr" and "adam_lr" for muon optimizer
delock Oct 30, 2025
b813c93
Merge branch 'master' into gma/muon_sep_lr
sfc-gh-truwase Nov 3, 2025
df0d913
Merge branch 'master' into gma/muon_sep_lr
delock Nov 5, 2025
c2b7c2b
Merge branch 'master' into gma/muon_sep_lr
delock Nov 7, 2025
65bc485
Merge branch 'master' into gma/muon_sep_lr
delock Nov 11, 2025
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
14 changes: 10 additions & 4 deletions deepspeed/runtime/engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1599,15 +1599,21 @@ def _configure_basic_optimizer(self, model_parameters):
param_groups = []
if muon_params:
accepted_parameters = dict()
for key in ["lr", "momentum", "weight_decay"]:
for key in ["lr", "momentum", "weight_decay", "muon_lr"]:
Copy link
Collaborator

@PKUWZP PKUWZP Oct 30, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does the user need to specific "muon_lr" for their training jobs?

Copy link
Collaborator Author

@delock delock Oct 31, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not mandatory, 'lr' will apply to both muon and adam. However, 'muon_lr' will be used if present. Same for 'adam_lr'. Thinking them as advanced user setting.

if key in optimizer_parameters:
accepted_parameters[key] = optimizer_parameters[key]
if key == "muon_lr": # muon_lr will override lr
accepted_parameters['lr'] = optimizer_parameters[key]
else:
accepted_parameters[key] = optimizer_parameters[key]
param_groups.append(dict(params=muon_params, use_muon=True, **accepted_parameters))
if non_muon_params:
accepted_parameters = dict()
for key in ["lr", "betas", "eps", "weight_decay"]:
for key in ["lr", "betas", "eps", "weight_decay", "adam_lr"]:
if key in optimizer_parameters:
accepted_parameters[key] = optimizer_parameters[key]
if key == "adam_lr": # adam_lr will override lr
accepted_parameters['lr'] = optimizer_parameters[key]
else:
accepted_parameters[key] = optimizer_parameters[key]
param_groups.append(dict(params=non_muon_params, use_muon=False, **accepted_parameters))
optimizer = MuonWithAuxAdam(param_groups)
else:
Expand Down