Hey guys, I have a question regarding a query pipeline on an agent. Can the agent used be a ContextRetrieverOpenAIAgent? In the examples on the documentation they always use a ReAct agent. Does anybody know if this is possible? Is there any documentation regarding this?

Hey! Does anyone know if a query pipeline can take different paths or routes via an if condition or similar? Or does it have to be done on the whole set of functions?

Hello. I'm trying to output an intermediate step of a query pipeline into one output of the same query pipeline. Do you know if this is possible? If yes, may you refer me to an example? Not the streaming one. I didn't understood that one 😦

Hello. I'm trying to replicate the simpledirectoryReader example and vectorstoreindex example (the paul graham essay one) with my openai api key. However, I get this error: BadRequestError: Error code: 400 - {'error': {'message': "This model's maximum context length is 8192 tokens, however you requested 16723 tokens (16723 in your prompt; 0 for the completion). Please reduce your prompt; or completion length.", 'type': 'invalid_request_error', 'param': None, 'code': None}}. Do you know if there's a modification or something i need to do in open ai for it to run the vectorstoreinde

Hello. I'm trying to replicate the simpledirectoryReader example and vectorstoreindex example (the paul graham essay one) with my openai api key. However, I get this error: BadRequestError: Error code: 400 - {'error': {'message': "This model's maximum context length is 8192 tokens, however you requested 16723 tokens (16723 in your prompt; 0 for the completion). Please reduce your prompt; or completion length.", 'type': 'invalid_request_error', 'param': None, 'code': None}}. Do you know if there's a modification or something i need to do in open ai for it to run the vectorstoreindex?

Hello. Does anyone has tried to put memory on a query pipeline without using agent handlers? If yes… is there a tutorial on something that might work on this ? My query pipeline consists of multiple functions and retrievers. I tried to transform it to an agent handled pipeline but I was unable to do that.

Hello guys. I've been working recently using AzureOpenAI service on my LLM in LlamaIndex and trying to create a custom agent. Given that, I followed the tutorial found on https://docs.llamaindex.ai/en/stable/examples/agent/custom_agent.html, changing only the llm part for AzureOpenAI instead of OpenAI. When getting to the end of the tutorial, once I'm gonna "initialize" the agent, I got the following error (Shown on thread):