Feature spotlights

Distributed as easy to install binaries

A binary distribution ensures simple installation and distribution of the latest updates via a repository for Ubuntu Linux.

Includes most ML/DL frameworks

PowerAI is built with optimized versions of leading frameworks including: Caffe-bvlc, Caffe-ibm, Caffe-nv, Chainer, DIGITS, Torch, Theano, and TensorFlow.

Includes all dependencies and libraries

Built with the hundreds of dependencies and the optimal parallel libraries for deep learning performance.

Easy updates: code updates arrive via repository

No more reinstallation from scratch! PowerAI enables smooth updates for every component from an IBM repository.

Validated deep learning platform with each release

Seamless update experience—each release is tested on supported hardware and software configurations for functionality & performance.

Built for superior interfaces to unlock larger DL datasets

PowerAI takes advantage of the CPU:GPU NVLink interconnect to help support and load larger deep learning models than ever before. Train datasets that could never be trained before.

Dedicated support teams for deep learning

Help via phone and email is available from deep learning experts.

Flexibility to run additional applications

PowerAI runs on IBM Power Systems S822LC for HPC, a platform that runs not only your deep learning but also a wide variety of HPC and High Performance Data Analytics workloads.

Technical details

Software requirements

Ubuntu 16.04 LTS

    Hardware requirements

    IBM Power Systems S822LC for High Performance Computing (8335-GTB) is the hardware platform built for PowerAI.

    • 2 POWER8 CPUs
    • 128GB of memory, or greater, recommended
    • NVIDIA® Tesla® P100 with NVLink GPUs strongly recommended
    • NVIDIA NVLink interface to Tesla GPUs strongly recommended

    You may also be interested in

    IBM Power System S822LC for High Performance Computing

    For maximum performance in cluster deployments.

    Learn more

    IBM Spectrum LSF Suites

    Complete workload management for demanding HPC environments

    Learn more

    Try it now

    Learn more