Pull Requests pytorch/xla

Add a link of pytorch/xla tutorial video to the contribution page.

#9731 opened 2025-12-15 18:06 by iwknow

Optimize as_strided_copy fast path to support offset

#9728 opened 2025-12-10 13:18 by Zantares

Expose torch_xla.runtime.clear_computation_cache binding to clear computation cache

#9720 opened 2025-11-24 18:53 by jameszianxuTT

Fix typo in warning message for SPMD mode

#9718 opened 2025-11-20 21:09 by ddilbazTT

implement `XLAHooks` and register it to PyTorch when loaded.

#9683 opened 2025-10-20 03:32 by iwknow

Ensuring PJRT Client destroy/destructor is called

#9675 opened 2025-10-07 03:19 by saarthak-aws

Add destructor for XLAGraphExecutor to free resource

#9636 opened 2025-09-11 11:06 by Zantares

Test ci

#9633 opened 2025-09-10 09:46 by bhavya01

Adding functionality for metadata communication across hosts

#9570 opened 2025-08-19 23:06 by kvshbg-aws

xla: relax error message assertion in dot_general test (allow backend…

#9557 opened 2025-08-13 02:58 by Rahulchaube1

Added Multi-Host TPU tutorial

#9507 opened 2025-07-24 09:39 by vfdev-5

XLAShardedTensor.to_local() support

#9505 opened 2025-07-24 00:19 by Hoomaaan

Add tests for DTensor.from_local() support for XLA device mesh

#9504 opened 2025-07-24 00:05 by Hoomaaan

Support SPMD placeholder tensors

#9489 opened 2025-07-17 21:14 by rpsilva-aws

Update torchprime tests to Python 3.12 bug testing

#9481 opened 2025-07-15 17:59 by pgmoka

feat: abstraction of xla::OpSharding proto using wrapper class

#9467 opened 2025-07-10 00:29 by kvshbg-aws

Introduce multi-operand collective permute

#9450 opened 2025-07-08 00:37 by rpsilva-aws

Convert some XLA_CHECKs to fatal errors.

#9439 opened 2025-07-02 15:37 by zhanyong-wan

Add a test on OOM error

#9414 opened 2025-06-26 21:14 by zhanyong-wan

[wip]ruff inter on torchax

#9396 opened 2025-06-20 18:08 by zzzwen

[WIP] Add autocast in torchax

#9361 opened 2025-06-16 02:07 by qihqi

Clarify `xr.device_type()` API and use them

#9343 opened 2025-06-11 23:23 by ghpvnist

wip: use torch from a wheel

#9340 opened 2025-06-11 21:05 by rickeylev

Test making this work from forks

#9335 opened 2025-06-11 06:23 by tengyifei

Do not inherit secrets

#9331 opened 2025-06-10 22:30 by tengyifei

Update tuned block table and add benchmarking/autotune scripts for quantized matmul pallas kernel.

#9322 opened 2025-06-10 02:24 by vanbasten23

Add DLRM DCN v2 with JAX Sparsecore

#9287 opened 2025-06-03 20:01 by chandrasekhard2

WIP - do not review; Support python 3.12

#9274 opened 2025-06-01 16:24 by zhanyong-wan

Migrate `torch_xla.device()` to `torch.device('xla')`

#9253 opened 2025-05-27 22:03 by ghpvnist

Switch bazel to default to Clang for C/C++ compilation

#9169 opened 2025-05-14 18:26 by zhanyong-wan