Intel® AI DevCloud

Free cloud compute is available for Intel® AI Academy members. Use Intel® AI DevCloud powered by Intel® Xeon® Scalable processors for your machine learning and deep learning training and inference compute needs.

Request Access

Professors: Get access to the Intel AI DevCloud for your classroom. Apply now.


  • 30 days of free access
  • Extended access (up to six months) with an approved research proposal
  • 200 GB of file storage
  • Access to a remote cluster of Intel® Xeon® Scalable processors


Our team monitors the community forum Monday through Friday, 9:00 a.m. – 5:00 p.m., Pacific daylight time.

AI Academy Forum

Available Frameworks and Tools

  • neon™ framework
  • Intel® Optimization for Theano*
  • Intel® Optimization for TensorFlow*
  • Intel® Optimization for Caffe*
  • Intel® Distribution for Python* (including NumPy, SciPy, and scikit-learn*)
  • Keras* library



Review these high-level questions and answers to get started with the Intel AI DevCloud and AI.

How can I improve deep learning performance when my framework is running in the background?
In some situations, deep learning code with default settings does not take advantage of the full compute capability of the underlying machine on which it runs, especially when the code runs on Intel® Xeon® Scalable processors. Intel created optimization techniques that enable optimal CPU performance for popular frameworks like Caffe* and TensorFlow*. For more information, see Tips to Improve Performance for Popular Deep Learning Frameworks on CPUs.
Who can request access to the Intel AI DevCloud?
Developers, data scientists, professors, students, start-ups, and others who are members of Intel® AI Academy are eligible to request access.
How do I become a member of the Intel® AI Academy?
You can join the Intel AI Academy by requesting access or become a member by registering here.
What happens once I have received access?
Once you gain access, you will log on to a Linux*-based head node of a batch farm. There, you can stage your code and data, compile it, and then submit jobs to a queue. Once the queued job completes, your results will be in your home ($HOME) directory.
  • Jobs are scheduled on Intel® Xeon® Scalable processors.
  • Each processor has 24 cores with two-way hyperthreading.
  • Each processor has access to 96 GB of on-platform RAM (DDR4).
  • Only one job will run on any processor at a time.
  • You will get 200 GB of file storage quota.
  • Your home directory is not visible to other users.

Note: Once your access period expires, your home directory on the cluster will be deleted.

I don’t live in the United States. Can I still get access to the Intel AI DevCloud?
The Intel AI DevCloud is available to all members of the Intel AI Academy, and is accessible from any country.
How much do I have to pay to use the Intel AI DevCloud?
Nothing. It is free to the members of Intel® AI Academy. Joining the Intel AI Academy is also free.

Executing Jobs

To help developers execute jobs on the Intel AI DevCloud, type the following answers into the command line in Linux*.

How do I set total wall time to the maximum on the Intel AI DevCloud?
In the command line, type one of the following:
  • Echo python <> -l walltime=24:00:00 | qsub
  • Add wall time in bash script before running in qsub mode:#!/bin/sh#PBS -l walltime=24:00:00 python
How do I query a job?
In the command line, type:

qpeek -o <JOBID>

If this doesn't return any results, it is possible that the job is complete and qpeek did not get a chance to peek.

Alternatively run qsub with the "-k oe" option:

qsub -k oe my_script

Standard input/output and error will be dumped into your home directory. You can check at any time while jobs are running.

How do I increase the wall time?
In the command line, type one of the following:
  • #PBS -l walltime=<10:30>,mem=320kb
  • echo sleep 1000 | qsub -l walltime=<00:30:00>
How do I run several jobs on multiple nodes?
In the command line, type:

#PBS -l nodes=1:knl

How do I get the full information about a job?
In the command line, type:

qstat -f <JOBID>

How do I delete a job?
In the command line, type:

qdel <JOBID>

How do I request specific nodes for features?
In the command line, type:

echo numactl -H | qsub -l nodes=1:knl:cache

How do I find the architecture and features of the compute nodes available to me?
Run the following command in the login node:


How do I log in to a compute node?
In the command line, type:

qsub –I

How do I get details of the nodes?
In the command line, type:

pbsnodes -a