Open achempak-polymer opened 7 months ago
cc @RobertCraigie can you take a look?
@achempak-polymer thanks for the report, do you have numpy installed?
@RobertCraigie yup I do, the latest version
@achempak-polymer can you share any more details?
I can't reproduce this against the main OpenAI API, both versions are taking about 0.5-1s with your inputs and the text-embedding-3-large
model.
Confirm this is an issue with the Python library and not an underlying OpenAI API
Describe the bug
I'm not absolutely certain if this is an issue with the Python library, but after upgrading from v0.28.0 to v1.10.0, we noticed a significant increase in latency (by about a factor of 4x) when requesting embeddings via an Azure OpenAI ada v2 deployment. This was confirmed in the Azure portal, where latency was indeed about 4x higher immediately after we deployed our service using the upgraded package. After downgrading back to v0.28.0, the issue resolved itself.
To Reproduce
Code snippets
This is how we query Azure with v0.28.0:
With v1.10.0:
OS
debian:bullseye-slim
Python version
Python v3.11.7
Library version
v1.10.0