I've been stuck on llama index 0.8.64 since the package for openai 1.0 got released because it didn't support azure open ai with SubQueryEngine -> API calls would fail. Was wondering if that issue is gone now.
Azure works fine with sub question query engine. But for whatever reason you need to use at least 2023-07-01-preview, and newer versions of the LLM model
The functional calling supported by 2023-07-01 is when I use stuff that deal with Tools(funciton calling) like the SubqueryEngine so if I just use a normal Retriever the API version and model wouldn't matter in the newer versions of LlamaIndex right?