# Chat with large data set of document - best approach?

**URL:** <https://discuss.huggingface.co/t/chat-with-large-data-set-of-document-best-approach/78106>\
**Category:** Beginners\
**Created:** [March 19, 2024, 1:11pm UTC](https://discuss.huggingface.co/t/chat-with-large-data-set-of-document-best-approach/78106 "2024-03-19T13:11:20Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ZZ92](https://avatars.discourse-cdn.com/v4/letter/z/b487fb/32.png) [@ZZ92](https://discuss.huggingface.co/u/ZZ92)\
**Post date:** [March 19, 2024, 1:11pm UTC](https://discuss.huggingface.co/t/chat-with-large-data-set-of-document-best-approach/78106/1 "2024-03-19T13:11:21Z")

</div>

Hi all;

I was wondering what is the best approach in case of large dataset of Documents : 500 Go to retrieve specific information based on a prompt ? embedding still enough for that volume of Data ?  
what can you suggest ?

Thank you