# Elastic Machine Learning need GPU?

**URL:** <https://discuss.elastic.co/t/elastic-machine-learning-need-gpu/359471>\
**Category:** Elasticsearch\
**Tags:** elastic-stack-machine-learning\
**Created:** [May 14, 2024, 1:42pm UTC](https://discuss.elastic.co/t/elastic-machine-learning-need-gpu/359471 "2024-05-14T13:42:29Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![S\_n\_Ngo\_Hoang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/s_n_ngo_hoang/32/123759_2.png) [@S\_n\_Ngo\_Hoang](https://discuss.elastic.co/u/S_n_Ngo_Hoang)\
**Post date:** [May 14, 2024, 1:42pm UTC](https://discuss.elastic.co/t/elastic-machine-learning-need-gpu/359471/1 "2024-05-14T13:42:29Z")

</div>

Hello everyone, I want to ask about the resources to use when running Elastic Machine Learning.  
Will Elastic's ML feature when I deploy on-premises require a GPU? Or do I just need CPU and RAM?

---

<div class="post-metadata">

**Author:** ![stephenb](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/stephenb/32/40856_2.png) [@stephenb](https://discuss.elastic.co/u/stephenb)\
**Post date:** [May 14, 2024, 2:22pm UTC](https://discuss.elastic.co/t/elastic-machine-learning-need-gpu/359471/2 "2024-05-14T14:22:35Z")

</div>

Hi @S_n_Ngo_Hoang  
Just CPU RAM and a little disk..

---

<div class="post-metadata">

**Author:** ![S\_n\_Ngo\_Hoang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/s_n_ngo_hoang/32/123759_2.png) [@S\_n\_Ngo\_Hoang](https://discuss.elastic.co/u/S_n_Ngo_Hoang)\
**Post date:** [May 14, 2024, 2:30pm UTC](https://discuss.elastic.co/t/elastic-machine-learning-need-gpu/359471/3 "2024-05-14T14:30:37Z")

</div>

Thanks. Do you have any information or articles that talk specifically and clearly about this issue? I'm researching but haven't found any documents

---

<div class="post-metadata">

**Author:** ![stephenb](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/stephenb/32/40856_2.png) [@stephenb](https://discuss.elastic.co/u/stephenb)\
**Post date:** [May 14, 2024, 3:02pm UTC](https://discuss.elastic.co/t/elastic-machine-learning-need-gpu/359471/4 "2024-05-14T15:02:34Z")

</div>

> [@S\_n\_Ngo\_Hoang](#):
>
> this issue?

Apologies What issue?

Here is top level documentation

> **[Machine Learning in the Elastic Stack \[8.13\] | Elastic](https://www.elastic.co/guide/en/machine-learning/current/index.html)**

For a starting point on specification I would look at The configurations we use in elastic Cloud. Those are great starting specifications to build your own on-prem.

> **[Elasticsearch Service AWS default provider instance configurations |...](https://www.elastic.co/guide/en/cloud/current/ec-default-aws-configurations.html)**

---

<div class="post-metadata">

**Author:** ![S\_n\_Ngo\_Hoang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/s_n_ngo_hoang/32/123759_2.png) [@S\_n\_Ngo\_Hoang](https://discuss.elastic.co/u/S_n_Ngo_Hoang)\
**Post date:** [May 14, 2024, 3:14pm UTC](https://discuss.elastic.co/t/elastic-machine-learning-need-gpu/359471/5 "2024-05-14T15:14:02Z")

</div>

Ok, thank you very much!