ExecuTorch adds Arm Cortex-M CPU backend using CMSIS-NN kernels
Overview
PyTorch says ExecuTorch now runs PyTorch models directly on Arm Cortex-M CPUs through a backend that uses Arm's CMSIS-NN kernels, so models no longer need conversion to another format.
The backend covers the Scalar, DSP and MVE/Helium capability tiers from a single workflow, and is available in the ExecuTorch repository.
On the Alif Ensemble E8, PyTorch reports that four MLPerf Tiny models each run in 21.5 ms or less, using under 90 KiB of RAM on one Cortex-M55 core. It also says the ExecuTorch core runtime is under 20 KiB of flash.
Written by AI from the articles below · updated Oct 15, 11:03 AM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- PyTorch BlogOfficialExecuTorch adds Arm Cortex-M CPU support, running PyTorch models directly on microcontrollers
AIPyTorch's ExecuTorch now runs models directly on Arm Cortex-M CPUs through a backend that uses Arm's CMSIS-NN kernels, removing the need to convert models to other formats. The backend covers Scalar, DSP and MVE/Helium tiers, and on the Alif Ensemble E8 four MLPerf Tiny models each run in 21.5 ms or less using under 90 KiB of RAM on one Cortex-M55 core. The ExecuTorch core runtime is under 20 KiB of flash.
Heat trend
Not enough continuous observations to show a trend yet.