Skip to content

Pull requests: google/XNNPACK

New pull request New
214 Open 9,433 Closed
214 Open 9,433 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Added broadcast fix for dim = 1.
#9956 opened Apr 13, 2026 by copybara-service bot Loading…
[RVV] add rvv f32 kernels for velu, vgelu, vapproxgelu, ibilinear, ppmm, qc8w-gemm
#9954 opened Apr 13, 2026 by velonica0 Loading…
2
Add mixed precision bf16,fp32 -> bf16/f32 subtract/multiply/divide kernels
#9953 opened Apr 13, 2026 by copybara-service bot Loading…
And neonbf16 unary convert kernels.
#9951 opened Apr 12, 2026 by copybara-service bot Loading…
Remove sde workflow
#9949 opened Apr 11, 2026 by copybara-service bot Loading…
Default scheduled_buffer store depth to 0.
#9947 opened Apr 10, 2026 by copybara-service bot Loading…
Add WASM RELAXED-SIMD F32 RMINMAX microkernels
#9940 opened Apr 10, 2026 by yolanda15 Contributor Loading…
[RVV] add rvv f32 kernels for vcos,vexp,vlog,vsigmoid,vsin,vtanh
#9926 opened Apr 9, 2026 by ken-unger Contributor Loading…
3
Add Neon f32-bf16-vcvt kernels
#9922 opened Apr 8, 2026 by GregoryComer Contributor • Draft
This CL introduces several enhancements to the dot product scheduling and parallelism logic:
#9916 opened Apr 8, 2026 by copybara-service bot Loading…
Optimize dot product scheduling and parallelism.
#9910 opened Apr 8, 2026 by copybara-service bot Loading…
math.h for RISCV scalar fp16 respect XNN_ENABLE_RISCV_FP16_VECTOR
#9904 opened Apr 7, 2026 by copybara-service bot Loading…
1
Fix integer overflow in allocation size calculations
#9860 opened Apr 3, 2026 by mohammadmseet-hue Contributor Loading…
Reduce windows GH actions runners from 32-core to generic windows-2022 machines
#9854 opened Apr 1, 2026 by copybara-service bot Loading…
Fix stack buffer overflows in NCHW reduce rewrite and ynnpack shim functions
#9851 opened Apr 1, 2026 by mohammadmseet-hue Contributor Loading…
1
Fix missing bounds checks in subgraph API functions
#9845 opened Apr 1, 2026 by mohammadmseet-hue Contributor Loading…
2 tasks done
Fix stack buffer overflows in ynnpack channelwise quantized tensor and reduce
#9843 opened Apr 1, 2026 by mohammadmseet-hue Contributor Loading…
4
Add CMake build for YNNPACK
#9803 opened Mar 27, 2026 by copybara-service bot Loading…
[gn] Fix component builds
#9785 opened Mar 26, 2026 by copybara-service bot Loading…
Add 2 bit SSE GEMM microkernels
#9752 opened Mar 23, 2026 by copybara-service bot Loading…
Add SIMD mask and select operations.
#9749 opened Mar 20, 2026 by copybara-service bot Loading…
VNNI microkernels disable XNN_FORCE_REALIZATION
#9719 opened Mar 18, 2026 by copybara-service bot Loading…
Replicate the scalar scaling factor for 4-bit tensor-wise quantized weights to the number of channels, as is done for 8-bit weights.
#9718 opened Mar 18, 2026 by copybara-service bot Loading…
Added deprecation warnings for fp16 flags in XNNPACK.
#9695 opened Mar 16, 2026 by copybara-service bot Loading…
Add a graph rewrite to fallback to fp32 when fp16 isn't supported.
#9686 opened Mar 13, 2026 by copybara-service bot Loading…
ProTip! no:milestone will show everything without a milestone.