# Implementing Custom BERT-Based Text Embedding Model for Semantic Search in Elasticsearch

**URL:** <https://discuss.elastic.co/t/implementing-custom-bert-based-text-embedding-model-for-semantic-search-in-elasticsearch/349434>\
**Category:** Elasticsearch\
**Created:** [December 15, 2023, 8:17am UTC](https://discuss.elastic.co/t/implementing-custom-bert-based-text-embedding-model-for-semantic-search-in-elasticsearch/349434 "2023-12-15T08:17:57Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ali\_Zare](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/ali_zare/32/130041_2.png) [@Ali\_Zare](https://discuss.elastic.co/u/Ali_Zare)\
**Post date:** [December 15, 2023, 8:17am UTC](https://discuss.elastic.co/t/implementing-custom-bert-based-text-embedding-model-for-semantic-search-in-elasticsearch/349434/1 "2023-12-15T08:17:57Z")

</div>

Hello everyone,

I'm exploring the possibility of setting up a custom text embedding model using the BERT architecture for semantic search within Elasticsearch. I'm curious if it's feasible to integrate a personalized text embedding model into Elasticsearch for semantic search purposes. Has anyone attempted or achieved this before? Any insights or guidance on this topic would be greatly appreciated.

Thank you!

---

<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:** [December 15, 2023, 4:10pm UTC](https://discuss.elastic.co/t/implementing-custom-bert-based-text-embedding-model-for-semantic-search-in-elasticsearch/349434/2 "2023-12-15T16:10:22Z")

</div>

Hi @Ali_Zare Welcome to the community.

Elasticsearch definitely supports Bring Your Own Model

Perhaps take a look here

> **[Compatible third party NLP models | Machine Learning in the Elastic Stack...](https://www.elastic.co/guide/en/machine-learning/current/ml-nlp-model-ref.html)**

Your model will need to adhere to one of the supported standards/architecture

---

<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:** [January 12, 2024, 4:11pm UTC](https://discuss.elastic.co/t/implementing-custom-bert-based-text-embedding-model-for-semantic-search-in-elasticsearch/349434/3 "2024-01-12T16:11:02Z")

</div>

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