forked from NVIDIA-NeMo/Megatron-Bridge
-
Notifications
You must be signed in to change notification settings - Fork 9
Pull requests: radixark/Megatron-Bridge
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[model] fix: handle separate KV normalization in GLM TileLang attention
bug
Something isn't working
#37
opened Sep 10, 2026 by
yushengsu-thu
Collaborator
Loading…
[Do not merge - still verifiying]: GLM-5.3-Flash (glm5_next) bridge with KDA / DSA-kpool / mHC and a PEFT sequence-parallel fix
#35
opened Sep 3, 2026 by
yushengsu-thu
Collaborator
Loading…
[model] fix: register both MTP submodule spellings in qwen3_next_bridge
#11
opened Jun 10, 2026 by
Zhichenzzz
Loading…
ProTip!
Updated in the last three days: updated:>2026-09-09.