From the examples/matmul_clamp_f32_qsi8d32p_qsi4c32p dir
$ mkdir -p build && cd build
$ cmake -DCMAKE_C_COMPILER=/path/to/aarch64-none-linux-gnu-gcc -DCMAKE_CXX_COMPILER=/path/to/aarch64-none-linux-gnu-g++ -DCMAKE_BUILD_TYPE=Release ../
$ mkdir -p build && cd build
$ cmake -DCMAKE_TOOLCHAIN_FILE=/path/to/android-ndk/build/cmake/android.toolchain.cmake -DANDROID_ABI=arm64-v8a -DANDROID_PLATFORM=30 -DCMAKE_BUILD_TYPE=Release ../
$ ./matmul_clamp_f32_qsi8d32p_qsi4c32p --threads <thread count>
Replace <thread count> with the number of worker threads to use for the matmul operation. Use --help to list every supported command-line argument.