Skip to content

Pull requests: google/XNNPACK

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
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

Add native BF16 sigmoid, square, and rsqrt kernels for neon + avx512
#10976 opened Aug 12, 2026 by JakeStevens Contributor Loading…
Added Grouped Dot operator.
#10974 opened Aug 12, 2026 by copybara-service Bot Loading…
F32-VCOS fix for catastrophic cancellation
#10973 opened Aug 12, 2026 by copybara-service Bot Loading…
packing: fix NULL deref in KleidiAI QS4 bias-substitute allocation
#10972 opened Aug 12, 2026 by destro4evr-rgb Contributor Loading…
Add the input to get_tensor_shape to the slinky func
#10970 opened Aug 11, 2026 by copybara-service Bot Loading…
Add support for split_k in dynamic and static cases
#10968 opened Aug 11, 2026 by copybara-service Bot Loading…
Support multi-dimensional gather in subgraph.
#10966 opened Aug 11, 2026 by copybara-service Bot Loading…
Repair shared workspace users after setup errors
#10960 opened Aug 11, 2026 by rootd4ddy Loading…
Support static shape deduction in reshape for ynnpack.
#10958 opened Aug 10, 2026 by copybara-service Bot Loading…
[gn] Roll DEPS for August 2026
#10953 opened Aug 10, 2026 by copybara-service Bot Loading…
Add KleidiAI Advanced SIMD PF16 6x32 GEMM kernel
#10943 opened Aug 7, 2026 by Colm-in-Arm Contributor Loading…
Fix integer overflow in qc2w/qc4w requantization scale allocation
#10934 opened Aug 5, 2026 by prasanna8585 Contributor Loading…
ProTip! Mix and match filters to narrow down what you’re looking for.