|
TraceML: always-on runtime visibility for PyTorch training
|
|
1
|
161
|
October 1, 2026
|
|
[Project Announcement] numgraph: A High-Performance C++ Numerical Core for Learnable Graphs (Looking for Maintainers!)
|
|
2
|
73
|
September 30, 2026
|
|
Possible resources leak in void ProcessGroupNCCL::shutdown()
|
|
1
|
105
|
September 30, 2026
|
|
LTX-2.5 on 16GB Mac Mini M4: My ComfyUI Solution Explained 🔥 How It Actually Works
|
|
0
|
61
|
September 29, 2026
|
|
DataLoader shared-memory unlink race on Python 3.14 only after midnight (Ubuntu 22.04, PyTorch 2.12.0+cu130)
|
|
0
|
47
|
September 28, 2026
|
|
PyTorch MPS Discussions
|
|
0
|
59
|
September 28, 2026
|
|
Ease development by running computations on remote GPU
|
|
10
|
7405
|
September 26, 2026
|
|
torch.distributions.kumaraswamy.Kumaraswamy generates samples outside its support
|
|
12
|
794
|
September 24, 2026
|
|
Active-speaker-driven video cropping: what the inference actually cost
|
|
0
|
58
|
September 23, 2026
|
|
Comparing raw MRI vs FreeSurfer-derived features for multimodal fusion with a small dataset
|
|
0
|
97
|
September 17, 2026
|
|
[RFC-0036] Zero-GC 64-Byte Cache-Aligned Flat Arena for Speculative Decoding Verification
|
|
0
|
77
|
September 17, 2026
|
|
Official wheel downloads return HTTP 403
|
|
3
|
201
|
September 16, 2026
|
|
[ROCm][MI200] linear_cross_entropy recomputed logits use backward-only FP16 alternate GEMM path
|
|
0
|
115
|
September 16, 2026
|
|
[Showcase] pyforker: A python framework for [core functionality]
|
|
1
|
117
|
September 14, 2026
|
|
ShrinkAI: knowledge distillation and compression package
|
|
0
|
86
|
September 14, 2026
|
|
Clifra: a differentiable Clifford algebra computation layer for PyTorch
|
|
0
|
71
|
September 14, 2026
|
|
Missing debug version binaries
|
|
1
|
109
|
September 11, 2026
|
|
WebGPU visual PyTorch debugger
|
|
2
|
165
|
September 11, 2026
|
|
Anker Innovations — Embodied AI / Agent Engineer — 2027 graduates, Shenzhen
|
|
0
|
148
|
September 10, 2026
|
|
PyTorch Symmetric Memory backend NCCL/NVSHMEM vs PyTorch Distributed backend NCCL
|
|
0
|
122
|
September 10, 2026
|
|
[Question] What is the relationship between PyTorch Symmetric Memory and Torchcomms?
|
|
0
|
86
|
September 9, 2026
|
|
[Maxpooling] what should the indices be when there're multiple max values?
|
|
4
|
1102
|
September 9, 2026
|
|
C10::bit_cast rejects c10::complex<float>/c10::complex<double>, stricter than std::bit_cast
|
|
1
|
141
|
September 8, 2026
|
|
1.43Ă— Faster MoE Training: LoongForge Redefines EP Expert Load Balancing with Topology-Aware Optimal Transport
|
|
0
|
136
|
September 8, 2026
|
|
C10 Runtime issue when building with VCPKG
|
|
2
|
130
|
September 8, 2026
|
|
Need help with an error
|
|
7
|
6012
|
September 7, 2026
|
|
What does “normal random sampling” mean in LLM text generation?
|
|
0
|
82
|
September 7, 2026
|
|
TypeError: Object of type int32 is not JSON serializable
|
|
2
|
3744
|
September 5, 2026
|
|
Cannot convert tensor_hash code to any integer
|
|
1
|
128
|
September 2, 2026
|
|
My index is out of bounds for some graph in gcn model
|
|
1
|
130
|
September 1, 2026
|