Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
204 Open 4,680 Closed
204 Open 4,680 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[WIP] phase2 basic shared ptr
#5960 opened Feb 12, 2026 by mdavis36 • Draft
2
skip reduction when comparing alloc domains in transpose scheduler
#5959 opened Feb 12, 2026 by liqiangxl Loading…
4
Fusion Statement Registration
#5958 opened Feb 12, 2026 by mdavis36 • Draft
2
Remove manual sharding in reshape tests
#5957 opened Feb 12, 2026 by Priya2698 Loading…
4
Ownership Fusion special types
#5954 opened Feb 11, 2026 by mdavis36 • Draft
5
Fusion val ownership
#5953 opened Feb 11, 2026 by mdavis36 Loading…
7
Llu/transpose tma
#5945 opened Feb 9, 2026 by liqiangxl • Draft
disable broadcast aliasing for the reduction + broadcast pattern
#5944 opened Feb 9, 2026 by liqiangxl Loading…
21
Use the default/cached value when a BuildConfig field is None
#5943 opened Feb 9, 2026 by wujingyue Loading…
1
6
skip reduction domain when compare allocation domain mapping in transpose scheduler
#5938 opened Feb 8, 2026 by liqiangxl Loading…
5
Clean up getContigMergeOfInnerSize
#5936 opened Feb 8, 2026 by wujingyue Loading…
12
remove input cached smem bank conflicts in transpose scheduler
#5930 opened Feb 6, 2026 by liqiangxl • Draft
7
increase transpose tile size to meet required bytes in flight
#5928 opened Feb 6, 2026 by liqiangxl • Draft
2
Broadcast-based allgather in host for-loop
#5925 opened Feb 6, 2026 by Priya2698 Loading…
26
Reduce-based MM+RS in MultiDeviceExecutor
#5923 opened Feb 5, 2026 by nsarka Loading…
@nsarka
35
[Ir Refactor] shared_ptr IrContainer
#5918 opened Feb 4, 2026 by mdavis36 • Draft
5
Remove legacy bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects.
#5917 opened Feb 3, 2026 by rdspring1 Loading…
23
Remove a deprecated/invalid argument
#5913 opened Feb 3, 2026 by wujingyue Loading…
5
Fusion axioms_ type.
#5908 opened Feb 2, 2026 by mdavis36 Loading…
1
4
Unnecessary header
#5897 opened Jan 29, 2026 by wujingyue Loading…
4
Remove cartesian_product and any_check
#5892 opened Jan 29, 2026 by zasdfgbnm • Draft
4
3D sharding for triangle updates
#5890 opened Jan 29, 2026 by wujingyue • Draft
1
Add kernel based alltoallv and cuda backend for MoE dispatch and combine
#5863 opened Jan 22, 2026 by samnordmann Loading…
12
Use symmetric memory in matmul+allreduce reference implementation
#5837 opened Jan 17, 2026 by wujingyue • Draft
5
avoid warp diverge in warp specialized kernel
#5830 opened Jan 15, 2026 by liqiangxl Loading…
6
ProTip! Follow long discussions with comments:>50.