# Errors when running Llama3 8b on executorch QNN backend

**URL:** <https://discuss.pytorch.org/t/errors-when-running-llama3-8b-on-executorch-qnn-backend/212689>\
**Category:** ExecuTorch\
**Created:** [November 8, 2024, 6:25am UTC](https://discuss.pytorch.org/t/errors-when-running-llama3-8b-on-executorch-qnn-backend/212689 "2024-11-08T06:25:35Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Jiawei\_Liu1](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/jiawei_liu1/32/73621_2.png) [@Jiawei\_Liu1](https://discuss.pytorch.org/u/Jiawei_Liu1)\
**Post date:** [November 8, 2024, 6:25am UTC](https://discuss.pytorch.org/t/errors-when-running-llama3-8b-on-executorch-qnn-backend/212689/1 "2024-11-08T06:25:35Z")

</div>

I was following the executorch tutorial sesion “Building and Running Llama 3 8B Instruct with Qualcomm AI Engine Direct Backend”. I’ve finished converting the llama3 8b and the llama\_main runner compiling. But I got this error when I was running the llama\_main runner on the phone through the command “./llama\_main --model\_path llama3\_8b.pte --tokenizer\_path tokenizer.model --prompt “What is the capital of France?” --seq\_len 120 --warmup=1”:

 ![image](https://discuss.pytorch.org/uploads/default/original/3X/d/8/d8ce86adb21ef9a98ae88fb31231c598efc6acdd.jpeg)  
I was pushing the model to meizu 21pro which is 8gen3 android phone with 16G RAM.  
Did anyone have this question before?

---

<div class="post-metadata">

**Author:** ![Jiawei\_Liu1](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/jiawei_liu1/32/73621_2.png) [@Jiawei\_Liu1](https://discuss.pytorch.org/u/Jiawei_Liu1)\
**Post date:** [November 8, 2024, 6:26am UTC](https://discuss.pytorch.org/t/errors-when-running-llama3-8b-on-executorch-qnn-backend/212689/2 "2024-11-08T06:26:53Z")

</div>

Additional error log

 ![image](https://discuss.pytorch.org/uploads/default/original/3X/1/6/16c4849fc383c074ed928319442a797a815e2f4e.jpeg)

---

<div class="post-metadata">

**Author:** ![Jiawei\_Liu1](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/jiawei_liu1/32/73621_2.png) [@Jiawei\_Liu1](https://discuss.pytorch.org/u/Jiawei_Liu1)\
**Post date:** [November 8, 2024, 6:27am UTC](https://discuss.pytorch.org/t/errors-when-running-llama3-8b-on-executorch-qnn-backend/212689/3 "2024-11-08T06:27:41Z")

</div>

Files in my phone directory:

 ![image](https://discuss.pytorch.org/uploads/default/original/3X/1/b/1b8bab21adf038e95950e762f57ba4674bb8a89e.png)

---

<div class="post-metadata">

**Author:** ![dbort](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/dbort/32/64698_2.png) [@dbort](https://discuss.pytorch.org/u/dbort)\
**Post date:** [December 2, 2024, 7:27pm UTC](https://discuss.pytorch.org/t/errors-when-running-llama3-8b-on-executorch-qnn-backend/212689/4 "2024-12-02T19:27:45Z")

</div>

Hi @Jiawei_Liu1, sorry for the delay. Could you please file a [github issue](https://github.com/pytorch/executorch/issues/new/choose) with instructions for reproducing the problem, along with information about your development environment?