-
Notifications
You must be signed in to change notification settings - Fork 76k
Pull requests: tensorflow/tensorflow
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Run Triton on cuda 13.2 and revert Triton patches that revert cuda 13-only changes in Triton
#125125
opened Aug 12, 2026 by
copybara-service
Bot
•
Draft
Reverts 2a435b40d1922821273bca94729556a2b38384f6
#125123
opened Aug 12, 2026 by
copybara-service
Bot
•
Draft
Add Linux and Windows build scripts and GitHub Actions CI workflow for TensorFlow CPU No-AVX builds.
#125122
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
tweak jax remat-checking logic for JAX_REMAT3=1 change
#125121
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix resource leak in Semaphore::ScopedReservation move assignment
#125120
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix type confusion vulnerability in AsyncValueTensor::FromTensor by validating root buffer and pointer alignment.
#125117
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
[Mosaic] Allow ND vectors in tpu.vector_store_idx
#125116
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
[XLA:CPU] Move opt barrier remover after cpu scheduler.
#125115
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix memory leak in GradientTape::ComputeGradient on backward function error.
#125114
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix resource leak and missing self-assignment check in Semaphore::ScopedReservation move assignment operator
#125113
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Add SparseCore Roofline Metrics support in backend and frontend. Include SparseCore Tile-Level (TEC) execution lines in OpMetricsDb extraction and differentiate async-done wait targets by associated async-call op name in HLO categorization.
#125112
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Allow tflite_extra_gles_deps usage for all Linux variants
#125111
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Support kTfLiteFloat16 for dynamically quantized fully connected op in XNNPACK delegate.
#125106
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix resource leak in Semaphore::ScopedReservation move assignment
#125105
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix type confusion in AsyncValueTensor::FromTensor
#125100
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Forward execution_stream_id through the IFRT IR program interpreter
#125099
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix resource leak and self-assignment bug in Semaphore::ScopedReservation move assignment operator.
#125098
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Feat: Add recursive=True support to tf.io.gfile.glob (fixes #39565)
awaiting review
Pull request awaiting review
size:L
CL Change Size: Large
#125096
opened Aug 11, 2026 by
sreeramakhil
Loading…
Make XLA LatencyHidingScheduler thread-safe for parallel simulation and evaluation.
#125095
opened Aug 11, 2026 by
copybara-service
Bot
•
Draft
Fix tf.reshape() to reject multiple -1 dims consistently
size:M
CL Change Size: Medium
#125093
opened Aug 11, 2026 by
AnupamKumar-1
Contributor
Loading…
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.