# Tensorflow binary not compiled to support AVX2 FMA

**URL:** <https://deeptalk.lambda.ai/t/tensorflow-binary-not-compiled-to-support-avx2-fma/267>\
**Category:** Uncategorized\
**Created:** [April 14, 2018, 3:22pm UTC](https://deeptalk.lambda.ai/t/tensorflow-binary-not-compiled-to-support-avx2-fma/267 "2018-04-14T15:22:42Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![sabalaba](https://sea1.discourse-cdn.com/flex019/user_avatar/deeptalk.lambda.ai/sabalaba/32/4_2.png) [@sabalaba](https://deeptalk.lambda.ai/u/sabalaba)\
**Post date:** [April 21, 2018, 6:35am UTC](https://deeptalk.lambda.ai/t/tensorflow-binary-not-compiled-to-support-avx2-fma/267/2 "2018-04-21T06:35:16Z")

</div>

One thing that I notice is that you are running out of memory (you can see that from the `CUDA_ERROR_OUT_OF_MEMORY` errors that you’re getting:

```auto
2018-04-14 08:01:57.171981: E tensorflow/stream_executor/cuda/cuda_driver.cc:936] failed to allocate 152.62M (160038912 bytes) from device: CUDA_ERROR_OUT_OF_MEMORY
2018-04-14 08:01:57.172873: E tensorflow/stream_executor/cuda/cuda_driver.cc:936] failed to allocate 137.36M (144035072 bytes) from device: CUDA_ERROR_OUT_OF_MEMORY

```

Can you try to re-run your job with a smaller batch size?

---

_[View the full topic](https://deeptalk.lambda.ai/t/tensorflow-binary-not-compiled-to-support-avx2-fma/267)._