Download Latest Version arm_compute-v53.2.0-android-aarch64-cpu-bin.tar.gz (23.4 MB)
Email in envelope

Get an email when there's a new version of Compute Library

Home / v53.1.0
Name Modified Size InfoDownloads / Week
Parent folder
NOTICE 2026-05-20 443.5 kB
arm_compute-v53.1.0-macos-aarch64-cpu-bin.tar.gz 2026-05-20 15.7 MB
arm_compute-v53.1.0-linux-armv7a-cpu-gpu-bin.tar.gz 2026-05-20 31.2 MB
arm_compute-v53.1.0-android-aarch64-cpu-bin.tar.gz 2026-05-20 23.4 MB
arm_compute-v53.1.0-android-aarch64-cpu-gpu-bin.tar.gz 2026-05-20 38.5 MB
arm_compute-v53.1.0-linux-aarch64-cpu-bin.tar.gz 2026-05-20 26.6 MB
arm_compute-v53.1.0-linux-aarch64-cpu-gpu-bin.tar.gz 2026-05-20 44.4 MB
arm_compute-v53.1.0-linux-armv7a-cpu-bin.tar.gz 2026-05-20 14.8 MB
README.md 2026-05-18 652 Bytes
v53.1.0 source code.tar.gz 2026-05-18 8.5 MB
v53.1.0 source code.zip 2026-05-18 14.6 MB
Totals: 11 Items   218.1 MB 1

v53.1.0 Public Release

Feat

  • Expose accumulation mode flag via Conv2dInfo
  • Add Bf16 SME matmul and reshape kernels
  • Enable BF16 I/O for CpuFullyConnected in the experimental Operator API

Fix

  • Default ScaleKernelInfo to no padding
  • Remove deprecated clang-tidy config option
  • Detection of HW caps and num Cpus for FreeBSD(R) and OpenBSD(R)

Performance

  • Enable signed requantized max pooling with differing qinfo
  • Enable assembly kernel for QASYMM8 MAX pooling

Documentation (API, build guide, contribution guide, errata, etc.) available here: https://artificial-intelligence.sites.arm.com/computelibrary/v53.1.0/index.xhtml

Source: README.md, updated 2026-05-18