Hey Team, how can I pass these values to a response object together with answer:
INFO:llama_index.token_counter.token_counter:> [get_response] Total LLM token usage: 1221 tokens
INFO:llama_index.token_counter.token_counter:> [get_response] Total embedding token usage: 0 tokens

another question, already asked, what is the best way to cost/performance point of view to ask/collect many questions answers?
I usually hit API rate limit, and huge bill

Could I set a fallback mechanism when I get using query engine router?

Are there any examples of usage of Contextual Compression Retriever with LLamaIndex?
https://python.langchain.com/en/latest/modules/indexes/retrievers/examples/contextual-compression.html?highlight=contextual%20compression

Hi LlamaIndex team, maybe you have an example of how I can use Ray with open-source models with Llama index?
https://www.anyscale.com/blog/llm-open-source-search-engine-langchain-ray
https://www.anyscale.com/blog/turbocharge-langchain-now-guide-to-20x-faster-embedding
https://www.anyscale.com/blog/building-a-self-hosted-question-answering-service-using-langchain-ray

To retrieve one question cost around 1000 tokens, is there a way to reduce the token usage? I already using optimiser

Hi!
Simple question. How can I chain query results or what is the best approach?
Example:
answer_1 = query_engine.query("Question 1?")
answer_2=query_engine.query("Question 2?")
answer_3=query_engine.query("Question 3?")

Make a report based on [answer_1, answer_2, answer_3].

Should I simply loop through questions/answers and feed them to LLMChain for the end result?
I tried to give all questions to the agent, but it finishes the chain after the first question is answered.
Also, agent some fail when the query router fails to select the engine for the query

how can I persist index, not to create every time?

Good morning! I have a question about sharing a knowledge base among users. I'd like to create a solution that stores all data in a single vector store or database. However, I want to restrict user access to only certain portions of the data. Is there a way to namespace the knowledge base so that each user can only access their designated areas?