|
Welcome to PyTorch Dev Forums
|
|
0
|
2750
|
January 22, 2021
|
|
Custom Ops Under torch.compile: autograd.Function vs torch.library.custom_op
|
|
4
|
144
|
May 18, 2026
|
|
Dropping Python 3.13t (free-threaded) support in nightlies and in the future PyTorch 2.13 release
|
|
0
|
45
|
May 14, 2026
|
|
PyTorch 2.12.0 General Availability
|
|
0
|
103
|
May 13, 2026
|
|
Inductor Passes
|
|
3
|
562
|
May 13, 2026
|
|
ABI Stability in the ML Ecosystem: LibTorch ABI Stable & TVM-FFI
|
|
0
|
73
|
May 12, 2026
|
|
Lead Core Maintainer Change
|
|
1
|
53
|
May 12, 2026
|
|
I keep getting this error while building PyTorch
|
|
3
|
89
|
May 8, 2026
|
|
LibTorch ABI Stable Plans 2026
|
|
0
|
103
|
May 7, 2026
|
|
PR Count increase as of May 2026
|
|
0
|
123
|
May 7, 2026
|
|
PyTorch 2.12 Final RC produced for pytorch & torchvision
|
|
0
|
256
|
May 5, 2026
|
|
Torch Documentation O3KR1 2026 H1
|
|
2
|
26
|
May 5, 2026
|
|
Making PyTorch Test Suite usable from out-of-tree braindump
|
|
1
|
110
|
April 27, 2026
|
|
Is PyTorch better than TensorFlow for beginners in AI/ML in 2025?
|
|
3
|
2351
|
April 24, 2026
|
|
Reminder — Calls for Features: PyTorch 2.12
|
|
0
|
225
|
April 20, 2026
|
|
[RFC] Complete uint16/uint32/uint64 Operator Coverage in ATen
|
|
1
|
72
|
April 20, 2026
|
|
Design question: supporting one-sided distributed GEMMs in DTensor
|
|
1
|
47
|
April 20, 2026
|
|
PyTorch 2.12 RC1 produced for pytorch & torchvision
|
|
0
|
237
|
April 16, 2026
|
|
How does the dispatcher work?
|
|
0
|
56
|
April 16, 2026
|
|
PyTorch Release 2.12 Branch Cut is Complete
|
|
0
|
74
|
April 16, 2026
|
|
Race conditions in persistent kernels
|
|
0
|
30
|
April 16, 2026
|
|
No PyTorch 2.11.1 release
|
|
0
|
107
|
April 14, 2026
|
|
RFC: Polyhedral Optimization Pass for PyTorch Inductor
|
|
1
|
178
|
April 13, 2026
|
|
Core Maintainer Meeting April 17th
|
|
0
|
48
|
April 11, 2026
|
|
DanLing NestedTensor on variable-length inputs: IMDB, model, and operator benchmarks
|
|
0
|
28
|
April 7, 2026
|
|
Torch.compile as a toolkit / manipulating dynamo outputs
|
|
4
|
156
|
April 5, 2026
|
|
Introducing CUDA 13.2 and Deprecating CUDA 12.8 (Release 2.12)
|
|
0
|
3836
|
April 1, 2026
|
|
Guard-Free Dynamic Shapes
|
|
0
|
64
|
March 31, 2026
|
|
Reducing Compile-Time Overhead in Unbacked-Symbol-Heavy torch.export Traces
|
|
0
|
26
|
March 31, 2026
|
|
Backed to Unbacked: From Guardable to Guardless Shapes in PyTorch
|
|
0
|
53
|
March 31, 2026
|