The knowledge base reader queries a permanent knowledge base containing your files, scraped URLs, and / or synced integrations (e.g., Google Drive).

Knowledge bases are important because they are able to retrieve the most relevant pieces of data based on a query. The most relevant data can then be passed to an LLM so that the LLM generates responses that are relevant to the data.

​
Node Inputs

  1. Search Query: The query that will be used to search the knowledge bases for relevant pieces.
    • Type: Text

​
Node Parameters

On the face of the node:

  1. Knowledge Base: Select the knowledge base you would like to query.
    • Type: KnowledgeBase

In the gear:

  1. Max chunks per query: the maximum number of pieces of data to be returned per query
  2. Enable Filter: enable the ability to write a metadata filter
  3. Enable Context: enable a text field to provide additional context for the search query
  4. Re-rank documents: Performs an additional reranking step to reorder the documents by relevance to the query
  5. Score Cutoff: The minimum relevancy score (between 0 and 1) that each piece of data will have semantically to the query
  6. Retrieval Unit: Return the most relevant chunks (text content) or Documents (will return document metadata)
  7. Transform Query: Transform the query for better results
  8. Answer Multiple Questions: Extract separate questions from the query and retrieve content separately for each question to improve search performance
  9. Expand Query: Expand query to improve semantic search
  10. Do Advanced QA: Use additional LLM calls to analyze each document to improve answer correctness
  11. Show Immediate Steps: Display the process the knowledge base is conducting at a given time in the chatUI
  12. Format Context for an LLM: Do an additional LLM call to format output

​
Node Outputs

If Retrieval Unit is set to Chunks

  1. Chunks: Semantically similar chunks retrieved from the knowledge base
    • Type: List<Text>
    • Example usage: {{knowledge_base_0.chunks}}

If Retrieval Unit is set to Documents

  1. Documents: Metadata for semantically similar documents retrieved from the knowledge base
    • Type: List<Text>
    • Example usage: {{knowledge_base_0.documents}}

If “Do Advanced QA” is enabled

  1. Response: A direct answer to the query
    • Type: Text
    • Example usage: {{knowledge_base_0.response}}

​
Considerations

  • Use a Knowledge Base Reader node when you want to query previously loaded data that has already been loaded. Use a Semantic Search node when your pipeline loads new data at run time for querying.
  • If the knowledge base is not returning relevant information to a query, try increasing the number of max chunks per query in the gear of the knowledge base.
  • For debugging purposes, you may attach an output node to the knowledge base to view the chunks that are returned for the query.

​
Example

The below example is a pipeline for chatting with a knowledge base. The knowledge base reader node returns chunks, which is utilized in the prompt of the LLM.

  1. Input Node: Represents the user message
  2. Knowledge Base Reader Node: Queries the knowledge base semantically
    • Search Query: {{input_0.text}}
  3. LLM Node: Responds to the user question
    • System (Instructions): Answer the question based on the context.
    • Prompt: Question: {{input_0.text}} Context: {{knowledge_base_1.chunks}}
  4. Output Node: Displays the response
    • Output: {{openai_0.response}}