# ARM-software/ComputeLibrary

The Compute Library is a set of computer vision and machine learning functions optimised for both Arm CPUs and GPUs using SIMD technologies.

Repository: https://github.com/ARM-software/ComputeLibrary
Canonical: https://ross.abutalabs.com/products/computelibrary
Language: C++
License Family: other
Topics: neon, opencl, computer-vision, arm, armv7, armv8, aarch64, machine-learning, simd, android, cpp, neural-network, sve, linux
Last push: 2026-08-12T09:47:06+00:00

## Health v2 (maintenance only)
Score: 96/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 91, longevity 100
- inputs: {"age_days": 3463, "days_push": 21, "days_rel": 63, "gap_med": 17.0, "n_releases_24m": 23}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3183, forks 818 (observed 2026-08-28T04:07:47.967371+00:00)

## What it is
Arm's Compute Library is a C++ collection of over 100 low-level machine learning and computer vision functions optimized for Arm Cortex-A/Neoverse CPUs and Mali GPUs using SIMD technologies like Neon, SVE2, and OpenCL. It provides multiple convolution algorithms, support for FP32/FP16/INT8/BFLOAT16 data types, and micro-architecture-tuned kernels with configurable lightweight builds.

## Use cases
- run neural network inference on arm cpus
- optimize convolutions with neon or sve2
- accelerate ml kernels on mali gpus with opencl
- deploy computer vision functions on android
- run inference on raspberry pi or embedded arm boards
- build lightweight ml binaries for bare metal arm

## When to choose
- you target arm cortex-a, neoverse, or mali hardware and need maximum performance
- you need immediate support for new arm technologies like sve2
- you want permissive MIT-licensed ml primitives with kernel fusion and tuning options

## When to avoid
- you need x86 or nvidia cuda as your primary target
- you want a high-level framework rather than low-level primitives
- you need a license file in-repo before legal review

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, computer-vision, image-processing, llm-inference, gpu-computing
- domain: machine-learning, computer-vision, deep-learning, embedded-systems
- platform: cpp, embedded
- tags: arm-neon, opencl, simd, sve, arm-mali, aarch64, kernel-fusion, bare-metal, linux, android, macos, gpu

## Member repositories
- ARM-software/ComputeLibrary (main) score 96

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:47.967371+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:25:01.738667+00:00, confidence not recorded.
  - readme: https://github.com/ARM-software/ComputeLibrary (fetched 2026-08-28T04:07:47.967371+00:00, sha 3b917f63b54e)
- Data as of 2026-08-30T08:39:29.467469+00:00.