# High query latency

**URL:** <https://community.pinecone.io/t/high-query-latency/6134>\
**Category:** Support\
**Tags:** vector-database, serverless\
**Created:** [July 15, 2024, 9:12am UTC](https://community.pinecone.io/t/high-query-latency/6134 "2024-07-15T09:12:43Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![waapi.machine.learni](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/waapi.machine.learni/32/3382_2.png) [@waapi.machine.learni](https://community.pinecone.io/u/waapi.machine.learni)\
**Post date:** [July 15, 2024, 9:12am UTC](https://community.pinecone.io/t/high-query-latency/6134/1 "2024-07-15T09:12:43Z")

</div>

We are experiencing very high latency, up to minutes when querying our index.

We are using:

- serverless index.
- dimensions 768.
- about 42 000 000 vectors with metadata.

We’re using the python sdk (grpc and rest gave same results)

It often occurs when no query were performed since some time. Our guess is, the delay is caused by the time to load the vectors like explain [here](https://community.pinecone.io/t/10x-slower-query-performance-from-lambda-func/4801/4). However, more than a few seconds make it completely unready for production.

---

<div class="post-metadata">

**Author:** ![patrick1](https://sea2.discourse-cdn.com/flex020/user_avatar/community.pinecone.io/patrick1/32/1863_2.png) [@patrick1](https://community.pinecone.io/u/patrick1)\
**Post date:** [July 15, 2024, 11:24am UTC](https://community.pinecone.io/t/high-query-latency/6134/2 "2024-07-15T11:24:57Z")

</div>

Hello @waapi.machine.learni welcome back to the forums.

I know you have an open Support case on this topic, so we will continue to work on it there.

I will post an update here for the community once we get to the bottom of the issue.