Skip to content

Pull requests: NVIDIA/Megatron-LM

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

Fix typo docs-only documentation only (docs or docstrings)
#4090 opened Apr 1, 2026 by janEbert Loading… Core 0.16
add permute fusion into hybrid ep complexity: low
#4089 opened Apr 1, 2026 by Autumn1998 Loading…
5 tasks
Set tensor-parallel attributes irrespective of perform_initialization complexity: low Final Review PR is in the "final review" stage
#4084 opened Mar 31, 2026 by ilml Loading…
5 tasks
Core 0.16
Part 1 Run functional tests
#4083 opened Mar 31, 2026 by asolergi-nv Draft
5 tasks
Core 0.16
[DEV] Minor update optimizer
#4082 opened Mar 31, 2026 by skyw Loading…
5 tasks
Core 0.16
Ko3n1g/chore/update nightlies
#4076 opened Mar 31, 2026 by ko3n1g Draft
5 tasks
Core 0.16
Fix THD RoPE offsets for local packed shards community-request Final Review PR is in the "final review" stage
#4075 opened Mar 31, 2026 by kaimo455 Loading…
3 of 5 tasks
Core 0.16
Enable FP8 DPA for MXFP8 recipe
#4066 opened Mar 30, 2026 by vasunvidia Draft
5 tasks
Preserve type of decorated methods/classes community-request Final Review PR is in the "final review" stage
#4062 opened Mar 30, 2026 by nschank Loading…
2 of 5 tasks
Fix forced lag initial conditions
#4056 opened Mar 30, 2026 by tdene Draft
5 tasks
ProTip! Updated in the last three days: updated:>2026-03-29.