Skip to main content

Retriever

A retriever in LlamaIndex is what is used to fetch Nodes from an index using a query string.

  • VectorIndexRetriever will fetch the top-k most similar nodes. Ideal for dense retrieval to find most relevant nodes.
  • SummaryIndexRetriever will fetch all nodes no matter the query. Ideal when complete context is necessary, e.g. analyzing large datasets.
  • SummaryIndexLLMRetriever utilizes an LLM to score and filter nodes based on relevancy to the query.
  • KeywordTableLLMRetriever uses an LLM to extract keywords from the query and retrieve relevant nodes based on keyword matches.
  • KeywordTableSimpleRetriever uses a basic frequency-based approach to extract keywords and retrieve nodes.
  • KeywordTableRAKERetriever uses the RAKE (Rapid Automatic Keyword Extraction) algorithm to extract keywords from the query, focusing on co-occurrence and context for keyword-based retrieval.
const retriever = vectorIndex.asRetriever({
  similarityTopK: 3,
});

// Fetch nodes!
const nodesWithScore = await retriever.retrieve({ query: "query string" });