Join open MLCommons task force on automation and reproducibility to participate in the collaborative development of the Collective Knowledge v3 playground (MLCommons CK) powered by the Collective Mind automation language (MLCommons CM) and Modular Inference Library (MIL) to automate benchmarking, optimization, design space exploration and deployment of Pareto-efficient AI/ML applications across any software and hardware from the cloud to the edge!

MLCommons Collective Knowledge v2 and v3

Grigori Fursin donated CM and CM4MLOps to MLCommons to benefit everyone. Since 2022, Collective Knowledge Technology is being developed by the public taskforce on automation and reproducibility at MLCommons:

Discontinued cTuning's Collective Knowledge v1 Technology (2015-2021):

CK-based Android apps

Collaborative AI/ML benchmarking using mobile devices: try it on your Android mobile phone
Collective training set with image misclassifications

CK browser plugin to annotate web pages with research objects

CK workflows

CK "solutions" for collaborative benchmarking and optimization of ML Systems

Modular and adaptive CK containers

All Docker containers with portable CK workflows

CK wrapper to simplify the user experience (cBench)

GitHub