# Using custom pytorch model in ES8

**URL:** <https://discuss.elastic.co/t/using-custom-pytorch-model-in-es8/308297>\
**Category:** Elasticsearch\
**Created:** [June 27, 2022, 9:54pm UTC](https://discuss.elastic.co/t/using-custom-pytorch-model-in-es8/308297 "2022-06-27T21:54:30Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![jyang](https://avatars.discourse-cdn.com/v4/letter/j/dec6dc/32.png) [@jyang](https://discuss.elastic.co/u/jyang)\
**Post date:** [June 27, 2022, 9:54pm UTC](https://discuss.elastic.co/t/using-custom-pytorch-model-in-es8/308297/1 "2022-06-27T21:54:30Z")

</div>

Hi,  
I would like to understand what it takes to deploy/use a custom pytorch model in elasticsearch 8. The desired usage would be to get text embedding for the input text from a custom SentenceTransformer model, using an elasticsearch api. Will probably deploy the model using eland.  
Is this doable? If it is, would this require the Platinum subscription for Elastic Cloud? I am currently trying this locally using a self managed version of ES8.  
Thanks.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [July 25, 2022, 9:55pm UTC](https://discuss.elastic.co/t/using-custom-pytorch-model-in-es8/308297/2 "2022-07-25T21:55:29Z")

</div>

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.