Forgive me if this is in the docs somewhere (I looked but couldn’t find), but are you able to confirm which OpenAI API endpoints DT4 uses? The reason I ask is that given my job I’m interested in enabling zero data retention, but that is only available for certain endpoints:
https://platform.openai.com/docs/guides/your-data#zero-data-retention
Further / alternatively, any chance of enabling the Azure OpenAI API?
Currently /v1/chat/completions is used but this might change in the future.
No such plans currently. There are just too many APIs and endpoints and we can’t support & test all of them.
Thanks for the swift response. Another, hopefully easier-to-implement question – could one manually configure the API base URL? Or at least let one choose the EU one?
See https://platform.openai.com/docs/guides/your-data#data-residency-controls:
“In Europe, you must also send requests to the https://eu.api.openai.com/ base URL for the request to be processed in the region.”
A future release might support this but no promises.
As OfflineLLM for macOS now offers a lokal server with OpenAI API, it would be great when DT4 allows us to use any compatible server.
As stated above, the request is noted with no promises.
(PS: These kind of apps/AI aggregators sprout and wither seemingly weekly. We can’t support every one and don’t have the time to install and test them all. Just something to consider.)
Thanks. That’s not a typical Aggregator. They use a custom implementation for GGUF models and claim to perform better than using Llama.cpp or using Apples MLX format instead.
Hm. This Person has some other apps out in the app store, all with very few to no reviews. The website offers nothing in way of information, lots of marketing blurb. And no address or such for the company.
All that makes me wary.
Indeed, but note I said apps/aggregrators 
