Skip to content

Issues: pytorch/xla

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

583 Open 1,751 Closed
583 Open 1,751 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
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Dataloading hangs on Trillium when using recent whls dataloading
#8407 opened Nov 21, 2024 by miladm
@bhavya01
Getting "undefined symbol: _ZN5torch4lazy13MetricFnValueB5cxx11E" with torch-xla nightly wheel for 2.6
#8406 opened Nov 21, 2024 by jeffhataws
@tengyifei
2
Einsum is not added to the supported list for autocast
#8405 opened Nov 21, 2024 by avizon-aws
2
Kaggle Notebook: model return loss None on TPU
#8402 opened Nov 20, 2024 by manh3152924
1
Add multiple TPU generations to the TPU CI tests CI CI related change tpuci usability Bugs/features related to improving the usability of PyTorch/XLA
#8399 opened Nov 19, 2024 by miladm
@lsy323
DDP doesn't synchronize weights
#8396 opened Nov 19, 2024 by darudimimi
torch.layer_norm causes RESOURCE_EXHAUSTED
#8395 opened Nov 19, 2024 by radna0
Prepare a subsection to educate users on the PyTorch workloads on AI-Hypercomputer documentation
#8389 opened Nov 18, 2024 by miladm
@mikegre-google
Need help validating TPU/XLA devices support for ComfyUI.
#8388 opened Nov 17, 2024 by radna0
1
Can Triton be used with XLA/TPU devices?
#8387 opened Nov 16, 2024 by radna0
1
How to write in-place custom ops compatible with torch.compile using pallas
#8385 opened Nov 15, 2024 by soodoshll
5
Support splitting physical axis in HybridMesh SPMD / Distributed
#8381 opened Nov 14, 2024 by tengyifei
How are PJRT asynchronous executions throttled by torch_xla?
#8380 opened Nov 14, 2024 by mcuiaws
7
Confusing text in bazel.md
#8379 opened Nov 13, 2024 by mikegre-google
@miladm @tengyifei @ManfeiBai
2
int8 StableHLO export Quantization
#8373 opened Nov 12, 2024 by Wheest
@lsy323
2
The precision error when xla parses irrational numbers leads to inconsistent calculation results with torch cuda
#8372 opened Nov 12, 2024 by mars1248
1
TPU Trillium Base Docker Image cannot initialize
#8371 opened Nov 12, 2024 by hsebik
6
xla bf16 compute accuracy issues
#8367 opened Nov 11, 2024 by mars1248
Export training model to StableHlo
#8366 opened Nov 8, 2024 by Zantares
2
Query regarding using 1 chip (2 cores of TPU v3) for Inference
#8359 opened Nov 6, 2024 by deepakkumar2440
Offer user guide instructions to users to leverage various libtpu versions documentation usability Bugs/features related to improving the usability of PyTorch/XLA
#8355 opened Nov 4, 2024 by miladm
@mikegre-google
1
Two different graph have the same hash value
#8353 opened Nov 4, 2024 by yitongh
1
RuntimeError: torch_xla/csrc/data_ops.cpp:185 : Check failed: input_sizes.size() <= output_sizes.size()
#8346 opened Oct 31, 2024 by samuele-bortolato1
Error when usecollective communication in torch_xla.core.xla_model.all_to_all in SPMD mopde
#8345 opened Oct 31, 2024 by DarkenStar
Bug - Using Sharding in Flash Attention with segment ids.
#8334 opened Oct 29, 2024 by dudulightricks
ProTip! Find all open issues with in progress development work with linked:pr.