You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It's a bit different actually, tho not enough to be much work. I somewhat recently refactored a chatbot UI lib (can't remember which) to support azure openai in addition to public chatgpt.
The biggest difference is that with public chatgpt, there's only one endpoint to connect to, and it accepts the model (and/or version, iirc) as a parameter for each call, whereas azure openai, each model/version is deployed separately, and so you have different endpoints (and thus different api keys, etc) for each model/version.
In practical terms, in that refactor I did, I had to generalize the endpoint config so that it assumed endpoint was an array of objects with their own api key and model and such, and in the models selector UX, make several api calls per endpoint etc, and then remember the model<->endpoint mapping for making the actual chat api call.
Not having looked at this codebase at all, I don't think this would be a super big lift.
I did not find way to connect Rubberduck to my Azure OpenAI deployment. Is it even supported? Thanks a lot.
The text was updated successfully, but these errors were encountered: