Skip to main content

RAG

RAG (Retrieval-Augmented Generation) components process a user query by retrieving relevant documents and generating a concise summary that addresses the user's question.

Vectara RAG

This component leverages Vectara's Retrieval Augmented Generation (RAG) capabilities to search and summarize documents based on the provided input. For more information, see the Vectara documentation.

Parameters

Inputs

NameTypeDescription
vectara_customer_idStringVectara customer ID
vectara_corpus_idStringVectara corpus ID
vectara_api_keySecretStringVectara API key
search_queryStringThe query to receive an answer on
lexical_interpolationFloatHybrid search factor (0.005 to 0.1)
filterStringMetadata filters to narrow the search
rerankerStringReranker type (mmr, rerank_multilingual_v1, none)
reranker_kIntegerNumber of results to rerank (1 to 100)
diversity_biasFloatDiversity bias for MMR reranker (0 to 1)
max_resultsIntegerMaximum number of search results to summarize (1 to 100)
response_langStringLanguage code for the response (e.g., "eng", "auto")
promptStringPrompt name for summarization

Outputs

NameTypeDescription
answerMessageGenerated RAG response

Hi, how can I help you?