Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
196 Open 4,313 Closed
196 Open 4,313 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

tma pointwise with broadcast
#5555 opened Nov 18, 2025 by liqiangxl • Draft
1
Extend the block quantization op to have optional swizzled block scales output
#5554 opened Nov 18, 2025 by protonu Loading…
31
add tma pointwise scheduler
#5553 opened Nov 18, 2025 by liqiangxl Loading…
23
Sequence parallel welford
#5552 opened Nov 18, 2025 by Priya2698 Loading…
12
[RFC] RaggedIterDomain for nested tensors
#5550 opened Nov 18, 2025 by naoyam • Draft
1
Add host IR LinearOut and lower LinearOp to that when needed
#5549 opened Nov 18, 2025 by wujingyue Loading…
7
Add PDL to schedulers
#5548 opened Nov 18, 2025 by rdspring1 • Draft
1
Inline Linear+Allreduce into the same loop
#5547 opened Nov 18, 2025 by wujingyue Loading…
11
[WIP] Use precompiled headers for nvFuser compilation
#5546 opened Nov 17, 2025 by jacobhinkle • Draft
12
Add more tma tests
#5540 opened Nov 15, 2025 by liqiangxl • Draft
1
Failed attempt to use OptOutMutator
#5538 opened Nov 15, 2025 by wujingyue • Draft
1
Mark host IR ops storing to preallocated outputs
#5537 opened Nov 15, 2025 by wujingyue • Draft
24
Inline communication and computation in the same loop
#5536 opened Nov 15, 2025 by wujingyue • Draft
1
3
[WIP] Support grouped GEMM in Cutlass+EVT
#5528 opened Nov 14, 2025 by jacobhinkle • Draft
1
Refactor host IR lowering
#5524 opened Nov 14, 2025 by wujingyue • Draft
claude pr pre-flight, github action version
#5523 opened Nov 14, 2025 by mattwalsh • Draft
@mattwalsh
[SymMem 5/5] Contiguous View
#5520 opened Nov 13, 2025 by samnordmann Loading…
3
[SymMem 4/5] Broadcast and Allgather lowering + runtime support
#5519 opened Nov 13, 2025 by samnordmann Loading…
4
[SymMem 3/5] Add MemoryType::Symmetric
#5518 opened Nov 13, 2025 by samnordmann Loading…
3
[SymMem 2/5] SymmetricTensor runtime type
#5517 opened Nov 13, 2025 by samnordmann Loading…
18
[SymMem 1/5] VMM and multicast reference
#5516 opened Nov 13, 2025 by samnordmann Loading…
14
Symmetric memory: full branch
#5515 opened Nov 13, 2025 by samnordmann • Draft
2
Remove redundant Set operation from broadcast when no broadcasting occurs
#5507 opened Nov 12, 2025 by Copilot AI • Draft
8 tasks done
1
@IvanYashchuk
17
Add Fusion.print, which prints the transforms as well as the math
#5499 opened Nov 11, 2025 by wujingyue Loading…
6
pointwise tma
#5498 opened Nov 11, 2025 by liqiangxl • Draft
3
ProTip! Type g i on any issue or pull request to go back to the issue listing page.