---
title: "Reranker"
slug: "reranker"
category: "glossary"
tags: ["glossary", "ai-agents", "rag", "ranking", "search"]
status: "stable"
last_updated: 2026-05-14
summary: "A reranker is a cross-encoder that scores query-document pairs after first-stage retrieval, improving precision over independent bi-encoder embeddings."
related:
  [
    "[[ai-agents/rag-retrieval]]",
    "[[ai-agents/rag]]",
    "[[glossary/embedding]]",
    "[[glossary/vector-similarity]]",
    "[[glossary/retrieval-augmented-generation]]",
    "[[ai-agents/embeddings]]",
  ]
---

## Overview

This page is the atomic definition. The RAG retrieval pipeline lives at [[ai-agents/rag-retrieval]].

## Definition

A reranker (cross-encoder) is a model that takes a query and a candidate document as a combined input and outputs a relevance score. This contrasts with bi-encoder [[glossary/embedding]] models, which encode query and document independently and compare their vectors. Cross-encoders are more accurate because they can attend to the interaction between query tokens and document tokens, but slower because they require a forward pass per candidate. The standard RAG pipeline runs reranking as a second stage: the first stage retrieves a broad candidate set (50-200 documents) using fast [[glossary/vector-similarity]] search; the reranker scores each candidate and returns the top 5-10 for insertion into the context. Popular reranking models: Cohere Rerank, Jina Reranker v2, Mixedbread mxbai-rerank, and cross-encoders from the Sentence Transformers library.

## When it applies

Add a reranker when retrieval precision is insufficient, meaning the top-k results include many irrelevant documents. Measure recall and mean reciprocal rank on an evaluation set before and after adding the reranker to confirm the improvement justifies the latency cost.

## Example

First-stage retrieval: 100 candidates via pgvector cosine search in 30 ms. Cohere Rerank scores all 100 in 80 ms and returns the 5 most relevant. The LLM receives 5 clean passages instead of 100, halving the context window cost.

## Related concepts

- [[ai-agents/rag-retrieval]] - the full retrieval pipeline including first-stage and reranking.
- [[glossary/embedding]] - bi-encoder models that power first-stage retrieval.
- [[glossary/vector-similarity]] - the metric used for first-stage retrieval that rerankers improve on.
- [[glossary/retrieval-augmented-generation]] - the RAG pattern that reranking sits inside.

## Citing this term

> See [[glossary/reranker|Reranker]] (llmbestpractices.com/glossary/reranker).

## Related

- [[ai-agents/rag-retrieval]]
- [[ai-agents/rag]]
- [[glossary/embedding]]
- [[glossary/vector-similarity]]
- [[glossary/retrieval-augmented-generation]]