# Does openai Question embeddings change everytime?

**URL:** <https://community.openai.com/t/does-openai-question-embeddings-change-everytime/970370>\
**Category:** API\
**Tags:** api-embedding\
**Created:** [October 7, 2024, 1:41pm UTC](https://community.openai.com/t/does-openai-question-embeddings-change-everytime/970370 "2024-10-07T13:41:06Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![merefield](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/merefield/32/14812_2.png) [@merefield](https://community.openai.com/u/merefield)\
**Post date:** [October 7, 2024, 2:36pm UTC](https://community.openai.com/t/does-openai-question-embeddings-change-everytime/970370/2 "2024-10-07T14:36:11Z")

</div>

Yes, very slightly, see this discussion:

> [@Can text-embedding-ada-002 be made deterministic?](https://community.openai.com/t/can-text-embedding-ada-002-be-made-deterministic/318054):
>
> I’m using text-embedding-ada-002 for creating semantic embeddings from paragraphs of text. However, each time I call the API with the same paragraph, I get slightly different vectors back. This is surprising, and actually not great, because it can generate unnecessary differences and non-determinism in downstream processes. Is there some kind of temperature parameter I can set to make it be deterministic? (none is documented in the API endpoint docs.) The text doesn’t change, the semantic of t…

That’s why you should use some kind of tolerance range to determine what is similar and what is not similar enough, the range giving you the necessary flex to include all things which are similar enough whilst also handling the case when something identical has slightly different numbers.

---

_[View the full topic](https://community.openai.com/t/does-openai-question-embeddings-change-everytime/970370)._