Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
185 Open 4,210 Closed
185 Open 4,210 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

Fix and Reenable Ring Allgather Cuda Ipc Test
#5429 opened Oct 24, 2025 by nsarka Loading…
@nsarka
1
Hacky patch to ensure nvfp4 for pointwise scheduler
#5428 opened Oct 23, 2025 by jjsjann123 Loading…
9
[prototype] CUDA graph capture/replay
#5427 opened Oct 23, 2025 by wujingyue • Draft
2
IdModel loop graph fix
#5426 opened Oct 23, 2025 by jjsjann123 Loading…
1
2
Escleanup
#5413 opened Oct 18, 2025 by zasdfgbnm • Draft
1
Refactor validation in test_repro.py Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Python API Issues related to the Python API
#5410 opened Oct 17, 2025 by rdspring1 Loading…
4
Enable greedy scheduler by default
#5406 opened Oct 17, 2025 by naoyam Loading…
5
Follow-up to #5353: Move another condition within views::filter
#5405 opened Oct 16, 2025 by Priya2698 Loading…
2
Llu/reshape
#5403 opened Oct 16, 2025 by liqiangxl • Draft
4
[Draft] Add support for block scaling factors with a swizzled output.
#5401 opened Oct 15, 2025 by protonu • Draft
1
LLVM include cxxabi bugfix.
#5396 opened Oct 15, 2025 by mdavis36 • Draft
3
[Inference Benchmark] Copy llama4 model code from transformers
#5388 opened Oct 14, 2025 by tbqh Loading…
5
Multicast tutorial
#5385 opened Oct 14, 2025 by samnordmann Loading…
4
Add print() function to fd.fusion printing to print full pre-scheduled fusion.
#5382 opened Oct 14, 2025 by kevinstephano Loading…
2
[multidevice][nvfp4 moe] Replicate weight not Shard & update desired inputs
#5376 opened Oct 13, 2025 by crcrpar • Draft
1
Port test_moe.py to multi-GPU
#5375 opened Oct 13, 2025 by wujingyue • Draft
1
Add support for automatic scheduler in direct bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Python API Issues related to the Python API
#5374 opened Oct 11, 2025 by rdspring1 Loading…
4
Grouping support of topk greedy scheduler
#5371 opened Oct 10, 2025 by naoyam • Draft
2
Use pointwise scheduler to schedule Block Quantization
#5362 opened Oct 9, 2025 by protonu • Draft
1
wjy/smoke
#5357 opened Oct 9, 2025 by wujingyue Loading…
2
save
#5354 opened Oct 9, 2025 by zasdfgbnm Loading…
2
Fix regiser alias cross for loops
#5350 opened Oct 8, 2025 by liqiangxl Loading…
6
Remove legacy python tests
#5341 opened Oct 7, 2025 by wujingyue Loading…
2
Remove test_moe.py which has been moved to Thunder
#5340 opened Oct 7, 2025 by wujingyue • Draft
4
Add fmin_fmax_promotion presegmentation pass
#5337 opened Oct 7, 2025 by tbqh Loading…
41
ProTip! Adding no:label will show everything without a label.