You try configuring the settings ?

and set your index llm
like
index_llm=" what ever llm is"
then query engines(llm=index_llm)

yes I defined the llm within the sittings. the llm specification is working within query engines generally, but not for the query engine made from the vectara managed index

i think vectera uses their own llm so you cant set it yourself its on their backend
@Logan M pls confirm

https://github.com/run-llama/llama_index/blob/07cb4c07409b0e0584e2dee70db0d3169f877b8e/llama-index-integrations/indices/llama-index-indices-managed-vectara/llama_index/indices/managed/vectara/retriever.py#L233

it does give this messages when I first load the index: LLM is explicitly disabled. Using MockLLM.
Embeddings have been explicitly disabled. Using MockEmbedding.

Dang sorry I should have stayed quiet 😅 but I know embedded model should be passed into the index, and you can always try to pass the Llm in into the index as well.

super appreciate your reply! I think I'm going to need to get creative if I want to do this