Workaround for 2 Second delay before processing API requests

[ulfkri]

Original Issue

I have been researching this issue I had earlier:

• 2 Second delay before processing API requests #373

I closed the issue because I wasn't able to reproduce it with curl. I'm using the official open ai nuget. I don't have the power to reopen the issue and comment on it, so I will post a workaround here:

Workaround

The two second delay disappears if I connect to http://127.0.0.1:1234/v1 in stead of http://localhost:1234/v1.

I'm sure localhost used to work fine with 0.2.x.

I will leave this issue open so you guys can find it and decide if it's an issue with LM Studio or with the open ai lib.

[yagil]

Thanks for the info, we'll look into it