Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
215 Open 4,496 Closed
215 Open 4,496 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

Add Github user for testing MCPs enable-auto-merge Auto-merge a PR when: 1) PR mergeable 2) Internal CI complete 3) No failures
#5778 opened Jan 8, 2026 by Steboss Loading…
2
GroupedBlockQuantizeOp PR2: Adding python API and updating llama4 benchmark
#5777 opened Jan 8, 2026 by jjsjann123 • Draft
1
GroupedBlockQuantizeOp PR1: Adding codegen support
#5776 opened Jan 8, 2026 by jjsjann123 Loading…
9
GroupedBlockQuantizeOp PR0: Adding runtime function
#5775 opened Jan 8, 2026 by jjsjann123 Loading…
8
Move class Scope from composite_nodes to internal_nodes enable-auto-merge Auto-merge a PR when: 1) PR mergeable 2) Internal CI complete 3) No failures
#5774 opened Jan 7, 2026 by wujingyue Loading…
3
Use meta device tensor to infer contiguity for expr-eval segments
#5772 opened Jan 7, 2026 by zasdfgbnm • Draft
1
Use toPolymorphicValue to convert py::handle directly to PolymorphicValue Direct Bindings Python extension with direct mapping to NvFuser CPP objects.
#5771 opened Jan 7, 2026 by rdspring1 Loading…
3
Add utility ir_utils::resetContiguityFromTensor
#5766 opened Jan 6, 2026 by zasdfgbnm • Draft
3
support dynamic shapes in warp specialized inner outer persistent scheduler
#5765 opened Jan 6, 2026 by liqiangxl Loading…
10
prioritze using smem only when cached inputs are persistent
#5764 opened Jan 6, 2026 by liqiangxl • Draft
9
skip unsupported arches in python tests
#5760 opened Jan 5, 2026 by liqiangxl • Draft
15
Add a toy multi-GPU benchmark
#5753 opened Jan 3, 2026 by wujingyue • Draft
13
Stream-parallelize loops
#5751 opened Jan 3, 2026 by wujingyue Loading…
1
48
[Build Speed][WIP] Dynamnic Type, Polymorphic Value, and Precompiled Headers
#5747 opened Jan 2, 2026 by csarofeen • Draft
4
CMake Dependency Validation & Reporting
#5740 opened Dec 26, 2025 by mdavis36 Loading…
52
[Build Speed] Dynamic type improvements
#5739 opened Dec 24, 2025 by csarofeen • Draft
2
Benchmark for nvfp4 scaled mm
#5737 opened Dec 23, 2025 by protonu Loading…
14
Prevent TMA usage for tensors with broadcast dimensions
#5735 opened Dec 22, 2025 by liqiangxl Loading…
5
Combine for RaggedIterDomain tensor list
#5716 opened Dec 19, 2025 by naoyam Loading…
10
Check consistency of multiplier
#5708 opened Dec 18, 2025 by zasdfgbnm • Draft
7
RaggedIterDomain cloning tensor list
#5707 opened Dec 18, 2025 by naoyam Loading…
11
Unconditionally return 1 for ParallelType::Stream for unshardedSizes
#5706 opened Dec 18, 2025 by zasdfgbnm • Draft
2
AllToAll Draft
#5705 opened Dec 18, 2025 by Priya2698 • Draft
1
Esdebug12
#5701 opened Dec 17, 2025 by zasdfgbnm • Draft
2
Add device type meta support for CutlassNvfp4GroupedMmaOp::evaluate
#5695 opened Dec 17, 2025 by zasdfgbnm Loading…
6
ProTip! Updated in the last three days: updated:>2026-01-05.