Collective Knowledge (CK) is an educational initiative to learn how to run AI, ML and other emerging workloads in the most efficient and cost-effective way across diverse models, data sets, software and hardware.

This community project was created by Grigori Fursin and is being developed in collaboration with MLCommons, MLPerf, FlexAI, cTuning, dedicated volunteers and contributors, and participants in open challenges.

You can learn more about this open-source project from the ACM TechTalk'21, Journal of Royal Society'20, ACM REP'23 keynote and our white paper.

Please visit the Collective Knowledge playground to dive deeper and learn more.