|
Backporting CVE fixes from Pytorch v2.6.0 to v2.1.2
|
|
0
|
278
|
May 12, 2025
|
|
Clarification of PyTorch Quantization Flow Support (in pytorch and torchao)
|
|
9
|
1388
|
May 9, 2025
|
|
PyTorch Release 2.7.1
|
|
0
|
3296
|
May 1, 2025
|
|
Conditional Decompositions
|
|
0
|
141
|
April 30, 2025
|
|
Using Nsight Systems to profile GPU workload
|
|
12
|
37627
|
April 30, 2025
|
|
Proposed Changes to Feature Tracking & Classification starting in Release 2.8
|
|
0
|
227
|
April 24, 2025
|
|
PyTorch Release 2.7.0 | Call for features
|
|
2
|
4413
|
April 23, 2025
|
|
PyTorch 2.7.0 General Availability
|
|
0
|
1278
|
April 23, 2025
|
|
Performance of "activation" sparsity
|
|
1
|
279
|
April 21, 2025
|
|
Why and how does aarch64 cpu wheel keep rpath instead of runpath?
|
|
1
|
200
|
April 18, 2025
|
|
New Contributor Looking for Mentorship!
|
|
5
|
611
|
April 17, 2025
|
|
CUDA Kernel with customised device
|
|
1
|
130
|
April 17, 2025
|
|
Intel GPU & CPU Enabling Status and Feature Plan – 2025 H1 Update
|
|
1
|
1575
|
April 16, 2025
|
|
Understanding CUDAGraph Trees
|
|
6
|
2037
|
April 15, 2025
|
|
Empty tensor with SymInt size
|
|
1
|
227
|
April 15, 2025
|
|
Possible to use custom backend to create TensorImpl that allows custom datatype?
|
|
2
|
525
|
April 11, 2025
|
|
Profiling torch.compile
|
|
0
|
299
|
April 10, 2025
|
|
PyTorch Release 2.7.0 - Final RC is available
|
|
0
|
3580
|
April 9, 2025
|
|
Float8 in PyTorch [1/x]
|
|
1
|
17792
|
April 7, 2025
|
|
Rename prune.remove
|
|
0
|
81
|
April 7, 2025
|
|
PyTorch 2.7 RC1 produced for pytorch, audio, vision
|
|
5
|
1557
|
April 7, 2025
|
|
What is the right way to use decompositions in dispatch mode?
|
|
0
|
239
|
April 5, 2025
|
|
Embrace tensor subclass as a Python device registration API
|
|
5
|
833
|
March 28, 2025
|
|
Inductor Triton Custom Op
|
|
6
|
2101
|
March 25, 2025
|
|
Higher Order Operators, 2023/10
|
|
2
|
3379
|
March 25, 2025
|
|
[BC Breaking] `torch.export.export_for_inference()` API is removed
|
|
0
|
152
|
March 24, 2025
|
|
How to prevent ops to be decomposed in custom compile backend
|
|
0
|
162
|
March 24, 2025
|
|
FSDP & CUDACachingAllocator: an outsider newb perspective
|
|
10
|
11401
|
December 13, 2024
|
|
Does PyTorch support RTX 5090?
|
|
1
|
2524
|
March 14, 2025
|
|
AOT module decompositions
|
|
1
|
465
|
March 13, 2025
|