🔍Retrivers

1)Cohere Rerank Retriever

Cohere Rerank indexes the documents from most to least semantically relevant to the query.

2)Embeddings Filter Retriever

A document compressor that uses embeddings to drop documents unrelated to the query.

3)HyDE Retriever

Use HyDE retriever to retrieve from a vector store.

4)LLM Filter Retriever

Iterate over the initially returned documents and extract, from each, only the content that is relevant to the query.

5)Prompt Retriever

Store prompt template with name & description to be later queried by MultiPromptChain.

6)Reciprocal Rank Fusion Retriever

Reciprocal Rank Fusion to re-rank search results by multiple query generation.

7)Similarity Score Threshold Retriever

Return results based on the minimum similarity percentage.

8)Vector Store Retriever

Store vector store as retriever to be later queried by Multi Retrieval QA Chain.

9)Voyage AI Rerank Retriever

Voyage AI Rerank indexes the documents from most to least semantically relevant to the query.

10)AWS Bedrock Knowledge Base Retriever

Purpose: Retrieves relevant documents or chunks from a Knowledge Base created and managed in Amazon Bedrock.

How It Works:

• Uses the Bedrock service's native vector store and retrieval tools.

• Typically paired with a Bedrock-supported embedding model (like Titan or Cohere).

• Use Case: Enterprise RAG applications hosted entirely on AWS infrastructure for scalability and security.

11)Custom Retriever

Purpose: Allows you to define your own logic for retrieving documents from a vector database or custom source.

How It Works:

• You can implement custom search logic (e.g., specific filters, hybrid retrieval, advanced ranking).

• Often used when built-in retrievers (like Pinecone or Qdrant) don't meet specific needs.

Use Case: When you need fine-grained control over how data is retrieved—e.g., combining metadata filtering, hybrid ranking, or integrating proprietary databases.

12)Jina AI ReRank Retriever

Purpose: Enhances standard retrieval by using Jina AI’s reranker models to re-score and reorder the results for better relevance.

How It Works:

• Performs an initial retrieval using embeddings or similarity search.

• Applies a reranker (often a cross-encoder model) to evaluate semantic similarity more accurately.

Use Case: Improves the quality of top-k results for question answering, especially in RAG workflows where the first retrieval might be too fuzzy

13)Multi Query Retriever

Purpose: Improves retrieval recall by expanding a single query into multiple semantically similar sub-queries.

How It Works:

• Generates multiple variations of the original question using an LLM.

• Sends all queries to the vector database and aggregates the results.

Use Case: Useful when user queries are ambiguous or underspecified—increases the chances of retrieving relevant data.