| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| arm_compute-v53.2.0-macos-aarch64-cpu-bin.tar.gz | 2026-07-01 | 15.7 MB | |
| NOTICE | 2026-07-01 | 443.5 kB | |
| arm_compute-v53.2.0-linux-armv7a-cpu-gpu-bin.tar.gz | 2026-07-01 | 31.2 MB | |
| arm_compute-v53.2.0-android-aarch64-cpu-bin.tar.gz | 2026-07-01 | 23.4 MB | |
| arm_compute-v53.2.0-android-aarch64-cpu-gpu-bin.tar.gz | 2026-07-01 | 38.5 MB | |
| arm_compute-v53.2.0-linux-aarch64-cpu-bin.tar.gz | 2026-07-01 | 26.6 MB | |
| arm_compute-v53.2.0-linux-aarch64-cpu-gpu-bin.tar.gz | 2026-07-01 | 44.4 MB | |
| arm_compute-v53.2.0-linux-armv7a-cpu-bin.tar.gz | 2026-07-01 | 14.8 MB | |
| README.md | 2026-06-30 | 607 Bytes | |
| v53.2.0 source code.tar.gz | 2026-06-30 | 8.5 MB | |
| v53.2.0 source code.zip | 2026-06-30 | 14.6 MB | |
| Totals: 11 Items | 218.2 MB | 1 | |
v53.2.0 Public Release
Feat
- Add experimental SME/SVE runtime selection controls
Fix
- Fix OpenCL(TM) C++ header include path in cl_definitions.h
- Enable beta OpenCL(TM) extension declarations for command buffers
- QuantizationInfo equality operator overload by treating implicit offset as zero
- Correct aarch64 preprocessor guard typo in DequantFP32_SupportedTypes test
Performance
- Speed up s8q NHWC max pooling
Documentation (API, build guide, contribution guide, errata, etc.) available here: https://artificial-intelligence.sites.arm.com/computelibrary/v53.2.0/index.xhtml