-
Notifications
You must be signed in to change notification settings - Fork 569
Pull requests: pytorch/xla
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Fix composite pass missing captured values in ops with nested regions
#9760
opened Mar 30, 2026 by
mstojkovicTT
Loading…
Fix persistent compilation cache not working on TPU
#9759
opened Mar 23, 2026 by
shr1ram
Loading…
3 tasks
Fix scan_layers to support layers with no weights
#9758
opened Mar 21, 2026 by
kredd2506
Loading…
2 tasks
Make a function wrapper for _get_xla_sharding_specs
#9757
opened Mar 17, 2026 by
ajakovljevicTT
Loading…
Expose torch_xla.runtime.clear_computation_cache binding to clear computation cache
#9720
opened Nov 24, 2025 by
jameszianxuTT
Loading…
implement
XLAHooks and register it to PyTorch when loaded.
#9683
opened Oct 20, 2025 by
iwknow
Collaborator
Loading…
Adding functionality for metadata communication across hosts
#9570
opened Aug 19, 2025 by
kvshbg-aws
Loading…
xla: relax error message assertion in dot_general test (allow backend…
#9557
opened Aug 13, 2025 by
Rahulchaube1
Loading…
Add tests for DTensor.from_local() support for XLA device mesh
#9504
opened Jul 24, 2025 by
Hoomaaan
Contributor
Loading…
Update torchprime tests to Python 3.12
bug
Something isn't working
testing
Testing and coverage related issues.
feat: abstraction of xla::OpSharding proto using wrapper class
#9467
opened Jul 10, 2025 by
kvshbg-aws
Loading…
Introduce multi-operand collective permute
#9450
opened Jul 8, 2025 by
rpsilva-aws
Collaborator
Loading…
Convert some XLA_CHECKs to fatal errors.
#9439
opened Jul 2, 2025 by
zhanyong-wan
Collaborator
•
Draft
Clarify
xr.device_type() API and use them
#9343
opened Jun 11, 2025 by
ghpvnist
Collaborator
Loading…
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.