# Build Apache TVM runtime for Adreno GPU on Android ARM64v8

**URL:** <https://discuss.tvm.apache.org/t/build-apache-tvm-runtime-for-adreno-gpu-on-android-arm64v8/17602>\
**Category:** Questions\
**Created:** [August 14, 2024, 4:46pm UTC](https://discuss.tvm.apache.org/t/build-apache-tvm-runtime-for-adreno-gpu-on-android-arm64v8/17602 "2024-08-14T16:46:05Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![varunnaw](https://discuss.tvm.apache.org/letter_avatar_proxy/v4/letter/v/ce73a5/32.png) [@varunnaw](https://discuss.tvm.apache.org/u/varunnaw)\
**Post date:** [August 14, 2024, 4:46pm UTC](https://discuss.tvm.apache.org/t/build-apache-tvm-runtime-for-adreno-gpu-on-android-arm64v8/17602/1 "2024-08-14T16:46:05Z")

</div>

I have two host platforms: a linux amd64 docker container and a macos m1 mac.

My target platform is an android device with a qualcomm soc: adreno gpu and hexagon dsp. I would like to just work with the adreno gpu for now.

How do I build tvm for this use case?

I am confused on how to even use tvm. I see python code in tutorials, but how do I get to that point? I’m obviously not going to run python on the android. How do I build tvm so that the runtime works for Android, but I can do model optimization on the host?

My goal is to create add opencl kernels as new operators that I can use in TVM.

Any help is greatly appreciated. I am in a serious time crunch.