-
Notifications
You must be signed in to change notification settings - Fork 75.3k
Pull requests: tensorflow/tensorflow
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[XLA:GPU] Remove TiledHloComputation from public API of the indexing cost model.
#117760
opened May 6, 2026 by
copybara-service
Bot
•
Draft
make compilation of llvm ir to ptx async for MlirKernelFusion
#117757
opened May 6, 2026 by
copybara-service
Bot
•
Draft
PR #40463: Use LLVM intrinsics for AllReduce CUDA Triton emitter (Enable ROCm Triton AllReduce 3/4)
#117756
opened May 6, 2026 by
copybara-service
Bot
•
Draft
PR #39302: Fix dynamic memcpy offset computation for host offloading with collective pipelining
#117754
opened May 6, 2026 by
copybara-service
Bot
•
Draft
Fully enable fusion of elementwise and other ops with reductions in YNNPACK
#117751
opened May 6, 2026 by
copybara-service
Bot
•
Draft
[XLA:MSA] Avoid constructing Shapes in Allocation::UpdateUses.
#117750
opened May 6, 2026 by
copybara-service
Bot
•
Draft
[XLA]Refactor HLO parser for async operations.
#117731
opened May 6, 2026 by
copybara-service
Bot
•
Draft
Support token types in
LinearizeHostBufferInto as well
#117730
opened May 6, 2026 by
copybara-service
Bot
•
Draft
[GPU] Enable RaggedAllToAll one-shot kernel with NCCL barrier for all GPU architectures.
#117729
opened May 6, 2026 by
copybara-service
Bot
•
Draft
PR #41901: Fix IFRT topology reporting device 0's device_kind for all devices
#117724
opened May 6, 2026 by
copybara-service
Bot
•
Draft
4 tasks done
2
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.