# Import from bucket when using integrated inference

**URL:** <https://community.pinecone.io/t/import-from-bucket-when-using-integrated-inference/7132>\
**Category:** Feature Requests\
**Tags:** vector-database\
**Created:** [December 11, 2024, 6:27pm UTC](https://community.pinecone.io/t/import-from-bucket-when-using-integrated-inference/7132 "2024-12-11T18:27:14Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![eldridge.henley](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/eldridge.henley/32/3618_2.png) [@eldridge.henley](https://community.pinecone.io/u/eldridge.henley)\
**Post date:** [December 11, 2024, 6:27pm UTC](https://community.pinecone.io/t/import-from-bucket-when-using-integrated-inference/7132/1 "2024-12-11T18:27:14Z")

</div>

Hi all.

Does the import from s3 work with this? If not, will there be an option for this in the future?

My current understand is that it does not due to the import files needing the embedding data within it.

Thanks and I appreciate everyones time.

---

<div class="post-metadata">

**Author:** ![gdj0nes](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/gdj0nes/32/223_2.png) [@gdj0nes](https://community.pinecone.io/u/gdj0nes)\
**Post date:** [December 11, 2024, 6:28pm UTC](https://community.pinecone.io/t/import-from-bucket-when-using-integrated-inference/7132/2 "2024-12-11T18:28:48Z")

</div>

Thanks for asking! We are actively looking into support this functionality. How many records would you be bringing in a typical import job and how often would it be running.

---

<div class="post-metadata">

**Author:** ![eldridge.henley](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/eldridge.henley/32/3618_2.png) [@eldridge.henley](https://community.pinecone.io/u/eldridge.henley)\
**Post date:** [December 11, 2024, 6:38pm UTC](https://community.pinecone.io/t/import-from-bucket-when-using-integrated-inference/7132/3 "2024-12-11T18:38:29Z")

</div>

Hello and thanks for the quick reply.

Looking at importing hundreds of millions of records initially and currently do not have an estimate on how often. But if it was able to ingest into existing namespaces, it would be deltas every hour or so.