DeepSpeed
Fix ZeRO-3 Muon all-gather for mixed parameter shapes
#8628
Merged

Fix ZeRO-3 Muon all-gather for mixed parameter shapes #8628

jinyouzhi
jinyouzhi Fix ZeRO-3 Muon all-gather for mixed parameter shapes
652bd20e
jinyouzhi jinyouzhi requested a review from tjruwase tjruwase 25 days ago
jinyouzhi jinyouzhi requested a review from tohtana tohtana 25 days ago
alanhuangyoo
alanhuangyoo commented on 2026-09-23
jinyouzhi
jinyouzhi Merge origin/master into fix-zero-3-muon-all-gather-for-mixed-par
08e34697
jinyouzhi Merge branch 'master' into fix_gloo_muon
234ce4a8
jinyouzhi Check full gradient shapes for ZeRO-3 Muon all-gather
caf356b9
jinyouzhi Merge remote PR updates into fix_gloo_muon
1aec7b23
alanhuangyoo
alanhuangyoo commented on 2026-09-23
jinyouzhi
pengdurice pengdurice added will-review
pengdurice Merge branch 'master' into fix_gloo_muon
be66d700
pengdurice
pengdurice approved these changes on 2026-09-28
pengdurice pengdurice enabled auto-merge 19 days ago
PKUWZP PKUWZP requested a review from PKUWZP PKUWZP 19 days ago
PKUWZP
PKUWZP approved these changes on 2026-09-28
pengdurice pengdurice merged c9e1b4d2 into master 19 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone