This shall guide you through the process of setting up a toolchain for the Kendryte K210 you may find on many well priced A.I. boards


For the development for a Kendryte K210 we use Linux (Ubuntu 20.04). As a recommendation, the installation in a virtual machine has proven itself here as well. Snapshots made it possible to restore the operating system to a defined state without losing much time. The use of Windows is possible but not covered in this article.

Compiler construction

You need to follow the guide on the Kendryte GitHub page. All requiered commands are given in that instuction.

SDK installation

The SDK is cloned with git clone https://github.com/kendryte/kendryte-standalone-sdk.git. With cd kendryte-standalone-sdk you can now change into the folder.
The SDK is based on CMake and is operated here with the help of the command lines. As a first test we try to compile the hello world project. With mkdir build && cd build we create the directory build and navigate into it. The hello_world provided with the SDK can now be compiled with cmake ... -DPROJ=hello_world -DTOOLCHAIN=/opt/kendryte-toolchain/bin. Afterwards the project can be compiled for the K210 with make.

Flash tools

Now that the first source code for the K210 has been successfully compiled, it's time to get it onto the board. For this you need kflash and the kflash_gui. First Python3 is needed and can be installed with sudo apt install python3-pip. kflash is now installed with sudo pip3 install kflash. For klash_gui a compiled version can be downloaded from this repository. After unpacking, a terminal can be used to change to the kflash_gui folder and ./kflash_gui to start the program.