Hi.. I need help in writing messages_to_prompt and completion_to_prompt for Mistral 7B. I tried the github, but its not clear. Any reference or guide is appreciated? Also do we need a separate output parser as well?

@WhiteFang_Jr Is there any support for Claude 3 models from bedrock api ? I can see that foundation models up to Claude 2 in aws bedrock integration

My set up is like this.
User sends a query to the RAG based chatbot (condense plus context one). There's a similarity cut off that doesn't pass irrelevant context to LLM. Is there any way, if the number of documents retrieved is zero, I can skip this call to LLM and return a generic response. @Logan M

How to get all nodes and embeddings saved in a vector store at any point in time ? Do we have any low level function for this?

@kapa.ai I want to use RAG over my data. I am using OpenSearch vector DB for vector store. I built an index and stored the content alongwith embeddings. Now I want to add some more documents to this, I used index.insert method and observed that it is stored in index. But I found that the new data is added in index but it is not added in vector store. I did index.refresh_ref_docs() as well. What am I missing here?

I want to understand in detail how the Simple directory reader works. Especially load_data, what arguments I can pass etc.?

Hi All. I searched the documents and found that there is no way to calculate time taken by each step in a RAG or Agent. So can you guide me how to calculate the time taken by each step such as retrieval, llm completion etc..

Can we run Evaluation pipeline without using OpenAI models? For example, Amazon Bedrock, llama2 , mistral etc?