# Bulk Metadata update feature

**URL:** <https://community.pinecone.io/t/bulk-metadata-update-feature/3930>\
**Category:** Feature Requests\
**Created:** [December 31, 2023, 4:25am UTC](https://community.pinecone.io/t/bulk-metadata-update-feature/3930 "2023-12-31T04:25:38Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![info4](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/info4/32/1618_2.png) [@info4](https://community.pinecone.io/u/info4)\
**Post date:** [December 31, 2023, 4:25am UTC](https://community.pinecone.io/t/bulk-metadata-update-feature/3930/1 "2023-12-31T04:25:38Z")

</div>

I want to update the metadata field ‘abcd’ for all vectors matching another metadata field value:  
filename=filename1

Suppose I query for all vectors with filename1 do one by one for each vector id as follows, it takes a long time:

index.update(id=match[‘id’], set\_metadata={“abcd”: new\_value})

Following doesnt work:  
index.update(ids=, set\_metadata={“abcd”: new\_value})

This also doesnt work:  
index.update(ids=, set\_metadata=[{“abcd”: new\_value} repeated same number of times as ids])

Can either of the approaches above be added as a feature?

---

<div class="post-metadata">

**Author:** ![andy-tmpt-me](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/andy-tmpt-me/32/1816_2.png) [@andy-tmpt-me](https://community.pinecone.io/u/andy-tmpt-me)\
**Post date:** [November 26, 2024, 7:29pm UTC](https://community.pinecone.io/t/bulk-metadata-update-feature/3930/2 "2024-11-26T19:29:21Z")

</div>

Would love bulk metadata update. I would be equally happy with a “batch” concept, similar to Firebase, just to not incur the networking overhead, something like:

batch = index.batch() // Batch interface supports a subset of index operations  
batch.update(id=X, set\_metadata…)  
batch.update(id=Y, set\_metadata…)  
batch.commit()