|
About the torchtitan category
|
|
0
|
170
|
September 9, 2024
|
|
CPU thread slow to enqueue GPU and communication kernels
|
|
2
|
36
|
October 20, 2025
|
|
How to apply selective activation checkpointing on _grouped_mm
|
|
0
|
45
|
September 20, 2025
|
|
[Distributed w/ TorchTitan] Breaking Barriers: Training Long Context LLMs with 1M Sequence Length in PyTorch Using Context Parallel
|
|
11
|
8485
|
August 29, 2025
|
|
Capture training graph with collectives via TorchTitan
|
|
8
|
180
|
August 15, 2025
|
|
Question about GPU memory usage when using pipeline parallelism training under larger micro batch count
|
|
4
|
121
|
July 30, 2025
|
|
[Distributed w/ TorchTitan] FLUX is Here: Experience Diffusion Model Training on TorchTitan
|
|
0
|
1066
|
June 27, 2025
|
|
Tensor parallel numeric mismatch
|
|
1
|
61
|
June 18, 2025
|
|
[Distributed w/ TorchTitan] Introducing Async Tensor Parallelism in PyTorch
|
|
11
|
15767
|
May 29, 2025
|
|
[Distributed w/ TorchTitan] Semi synchronous training using TorchFT
|
|
0
|
324
|
May 8, 2025
|
|
PyTorch Tensor Parallel
|
|
0
|
113
|
May 1, 2025
|
|
Dcp.save straight to cloud storage
|
|
5
|
217
|
April 15, 2025
|
|
How to avoid casting DTensor to Tensor before calling a custom operator (a CUDA kernel)
|
|
1
|
100
|
April 2, 2025
|
|
[Distributed w/ TorchTitan] Training with Zero-Bubble Pipeline Parallelism
|
|
0
|
3248
|
December 19, 2024
|
|
[Distributed w/ TorchTitan] Optimizing Checkpointing Efficiency with PyTorch DCP
|
|
0
|
3095
|
October 7, 2024
|
|
[Distributed w/ Torchtitan] Enabling Float8 All-Gather in FSDP2
|
|
0
|
2383
|
September 9, 2024
|