AIXPRT is a free benchmark tool for evaluating a system's machine learning interface performance by running common image-classification, object-detection, and recommended system workloads.


What does AIXPRT do?

Do I have to pay for this?

Where can I download AIXPRT?

Writing about AIXPRT

Are there restrictions on publishing results?

Can I use data from the AIXPRT results database in my article?

Running AIXPRT

What skills do I need to run AIXPRT?

What types of system can I run it on?

How long does it take to run?

Which operating systems does AIXPRT support?

What are the minimum hardware requirements?

Which hardware components can I target?

Which toolkits/workloads does AIXPRT run?

AIXPRT: the basics

The toolkits

The networks

The workloads

Can I test my system?

Testing and results

Running AIXPRT

Understanding AIXPRT results

The AIXPRT results database

Sharing your results

Key terms and concepts

Inference tasks (workloads)