Skip to content

Pull requests: NVIDIA/Fuser

New pull request New
177 Open 3,640 Closed
177 Open 3,640 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

Use driver API in getMaxActiveClusters
#4717 opened Jul 2, 2025 by jacobhinkle • Draft
2
use 255 if compute_branch_regs is not set
#4716 opened Jul 2, 2025 by liqiangxl Loading…
1
expose DataType::Byte, DataType::UInt16, and DataType::Uint32 to Python frontend
#4715 opened Jul 2, 2025 by crcrpar Loading…
[CI Run]
#4714 opened Jul 2, 2025 by Priya2698 • Draft
3
Minor refactoring of MoE gate tests
#4713 opened Jul 2, 2025 by naoyam Loading…
2
Improve fp8 casting
#4711 opened Jul 1, 2025 by zasdfgbnm Loading…
6
Move multidevice passes further to end
#4710 opened Jul 1, 2025 by Priya2698 • Draft
2
Set disable_replace_uniform to True for thunder benchmarks
#4709 opened Jul 1, 2025 by liqiangxl Loading…
4
[WIP] Check the propagation of reference tensors in pointwise scheduler
#4708 opened Jul 1, 2025 by jacobhinkle • Draft
1
1
Use loop domain in unshardedSizes for fusion inputs with DID parallelization
#4705 opened Jun 30, 2025 by Priya2698 Loading…
2
Add size, shape, define_vector, and reshape ops to direct bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Python API Issues related to the Python API
#4704 opened Jun 30, 2025 by rdspring1 • Draft
1
New preseg pass to re-use (expanded) broadcasts
#4702 opened Jun 30, 2025 by jacobhinkle • Draft
1
3
Add Linear Layer Support to Direct Python Bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Multi-GPU Thunder-Inference-Demo
#4701 opened Jun 28, 2025 by rdspring1 • Draft
1
Add support for matmul and linear ops to Direct Python Bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Python API Issues related to the Python API Thunder-Inference-Demo
#4698 opened Jun 28, 2025 by rdspring1 • Draft
1
Add support for cast operations to Direct Python Bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Python API Issues related to the Python API Thunder-Inference-Demo
#4697 opened Jun 28, 2025 by rdspring1 • Draft
1
Support vectorize cast for everything excluding fp8 and fp4
#4695 opened Jun 27, 2025 by zasdfgbnm • Draft
3
Use torch::kFloat4_e2m1fn_x2 for fp4
#4692 opened Jun 27, 2025 by zasdfgbnm Loading…
8
revise msg to originally detected persistent_buffers to avoid confusion
#4691 opened Jun 27, 2025 by liqiangxl Loading…
2
Vectorize embedding when fused with normalization operations
#4690 opened Jun 27, 2025 by liqiangxl Loading…
2
Add Multi-Gpu Support to Direct Python Bindings Direct Bindings Python extension with direct mapping to NvFuser CPP objects. Multi-GPU Python API Issues related to the Python API Thunder-Inference-Demo
#4689 opened Jun 27, 2025 by rdspring1 • Draft
18
NVFP4QuantizeTest.WithoutPerTensorAmax
#4688 opened Jun 27, 2025 by zasdfgbnm • Draft
1
Host IR LLVM Lowering 2: LaunchKernel node support
#4687 opened Jun 26, 2025 by wolfcomos • Draft
Add Cutlass NVFP4 Gemm to nvfuser_direct python bindings Cutlass Matmuls Thunder-Inference-Demo
#4676 opened Jun 26, 2025 by rdspring1 • Draft
18
FP4 support in pointwise scheduler
#4666 opened Jun 24, 2025 by zasdfgbnm • Draft
1
Create NVFP4 grouped gemm bindings in direct bindings Cutlass Matmuls Thunder-Inference-Demo
#4662 opened Jun 24, 2025 by rdspring1 • Draft
2 tasks
2
ProTip! Type g i on any issue or pull request to go back to the issue listing page.