Support load-balancing across OpenAI instances

[simonkurtz-MSFT]

Confirm this is a feature request for the Python library and not the underlying OpenAI API.

• This is a feature request for the Python library

Describe the feature or improvement you're requesting

Introduce the ability to use a prioritized set of backends and load-balance across different OpenAI instances.

Additional context

I wrote a Python OpenAI Load Balancer as an example of what may be useful.

[rattrayalex]

Thank you for sharing, @simonkurtz-MSFT ! cc @kristapratico

I don't expect we'll be able to add this to the library itself anytime soon, but I'll keep it in the back of my mind. In the meantime, will be happy to send folks towards your library.

[simonkurtz-MSFT]

Thank you! I have it as just a module, not a publishes library yet. That's a next step.

Do you all have a page that lists community projects for additional functionality that you link to?

[rattrayalex]

Not to my knowledge!