Collective Knowledge framework (CK) helps to organize software projects as a database of reusable components, automation recipes and portable workflows with common APIs and extensible meta descriptions based on FAIR principles. See the real-world use cases to support reproducible MLSys R&D, enable portable MLOps and automate ML/SW/HW co-design:





GitHub Stars



Last Commit

5d ago







Apache 2.0



Collective Knowledge framework (CK)

Downloads PyPI version Python Version

Linux/MacOS: Build Status Windows: Windows Build status

Colab Colab Documentation Status Coverage Status


Stable versions

The latest version of the CK automation suite supported by MLCommons™: v2.5.8 (Apache 2.0 license):

Development versions

We plan to develop a new version of the CK framework (v3) as a collaborative effort within different MLCommons workgroups - please feel free to join this community effort!

Deprecated versions

Versions 1.x including v1.17.0 and 1.55.5 (BSD license) are stable but not officially supported anymore. Please get in touch and we will help you to upgrade your infrastructure to use the latest MLCommons technology!


Collective Knowledge framework (CK) helps to organize software projects as a database of reusable components with common automation actions and extensible meta descriptions based on FAIR principles (findability, accessibility, interoperability and reusability) as described in our journal article (shorter pre-print).

Our goal is to help researchers and practitioners share, reuse and extend their knowledge in the form of portable workflows, automation actions and reusable artifacts with a common API, CLI, and meta description. See how CK helps to automate benchmarking, optimization and design space exploration of AI/ML/software/hardware stacks, simplifies MLPerf™ inference benchmark submissions and supports collaborative, reproducible and reusable ML Systems research:

Current projects



Follow this guide to install CK framework on your platform.

CK supports the following platforms:

As a host platformAs a target platform
Generic Linux
Linux (Arm)
Raspberry Pi
Bare-metal (edge devices)-±


Portable CK workflow (native environment without Docker)

Here we show how to pull a GitHub repo in the CK format and use a unified CK interface to compile and run any program (image corner detection in our case) with any compatible data set on any compatible platform:

python3 -m pip install ck

ck pull repo:mlcommons@ck-mlops

ck ls program:*susan*

ck search dataset --tags=jpeg

ck detect soft --tags=compiler,gcc
ck detect soft --tags=compiler,llvm

ck show env --tags=compiler

ck compile program:image-corner-detection --speed

ck run program:image-corner-detection --repeat=1 --env.MY_ENV=123 --env.TEST=xyz

You can check output of this program in the following directory:

cd `ck find program:image-corner-detection`/tmp


You can now view this image with detected corners.

Check CK docs for further details.

MLPerf™ benchmark workflows

Portable CK workflows inside containers

We have prepared adaptive CK containers to demonstrate MLOps capabilities:

You can run them as follows:

ck pull repo:mlcommons@ck-mlops
ck build docker:ck-template-mlperf --tag=ubuntu-20.04
ck run docker:ck-template-mlperf --tag=ubuntu-20.04

Portable workflow example with virtual CK environments

You can create multiple virtual CK environments with templates to automatically install different CK packages and workflows, for example for MLPerf™ inference:

ck pull repo:mlcommons@ck-venv
ck create venv:test --template=mlperf-inference-main
ck ls venv
ck activate venv:test

ck pull repo:mlcommons@ck-mlops
ck install package --ask --tags=dataset,coco,val,2017,full
ck show env

Integration with web services and CI platforms

All CK modules, automation actions and workflows are accessible as a micro-service with a unified JSON I/O API to make it easier to integrate them with web services and CI platforms as described here.

More examples

CK portal

We have developed the portal to help the community organize and find all the CK workflows and components similar to PyPI:

Testing CK internals and workflows

The community provides Docker containers to test CK and components using different ML/SW/HW stacks (DSE).


Users can extend the CK functionality via CK modules or external GitHub reposities in the CK format as described here.

Please check this documentation if you want to extend the CK core functionality and modules.

Note, that we plan to redesign the CK core to be more pythonic (we wrote the first prototype without OO to be able to port it to bare-metal devices in C but eventually we decided to drop this idea).

Author and coordinator


We would like to thank all contributors and collaborators for their support, fruitful discussions, and useful feedback! See more acknowledgments in the CK journal article.

Rate & Review

Great Documentation0
Easy to Use0
Highly Customizable0
Bleeding Edge0
Responsive Maintainers0
Poor Documentation0
Hard to Use0
Unwelcoming Community0