Complete the Reciprocal Rank Fusion (RRF) score formula

Attempt it

Reciprocal Rank Fusion combines two ranked retrieval results by summing for each candidate across retrievers, where k is a small constant (commonly 60) and rank is the candidate's in each retriever's result list.

Quick reference

TL;DR

RRF sums 1/(k+rank) over retrievers, where rank is the candidate's position in each retriever's result list and k smooths the curve.

Memory aid

Sign in for the mnemonic that makes this stick. See Pro pricing.

Easy to grasp

Picture two judges scoring contestants but on totally different scales. One scores out of ten, the other out of a hundred. You cannot just add their numbers; the bigger scale would always win. A fair fix is to ignore the raw numbers and only use the ranking. Each judge says who is first, second, third. You give every contestant a small bonus when a judge ranks them high, and a much smaller bonus when ranked low. Add the bonuses across judges and that is the final score. The bonus for rank position uses a soft slope so first place and second place do not differ by miles. That soft slope is the small number called k. Reciprocal Rank Fusion is exactly this trick applied to search results.

Key concepts

Embeddings

Detailed answer & concept explanation

~5 min read

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example. Click to expand.

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example.

Sign in for 5 free unlocks/day or go Pro for unlimited. See Pro pricing.

Quick scan · 2–4 min

Everything important, quickly.

60s: write the formula, name the two filled blanks, explain the `+ k` smoothing role, give the conventional `k=60` default, and connect it to hybrid retrieval as the rank-only fusion that avoids score calibration.

Real-world examples

Real products, models, and research that use this idea.

Elastic's hybrid retrieval API ships `rrf` as a named retriever with `k` defaulting to 60, paired with dense vector and BM25 sub-retrievers.
Vespa documents RRF as the canonical hybrid fusion with the exact `1/(k+rank)` formula and `k=60` default in its 2026 retrieval guide.
LangChain's EnsembleRetriever uses RRF under the hood when combining BM25 and embedding retrievers, exposing the same `k=60` default.
OpenSearch 2.x added a native RRF processor that fuses keyword and neural search results with the standard `1/(k+rank)` formula.

Interviewer follow-ups

What an interviewer would ask next. Try answering before peeking at the approach.

QWhy does the `+ k` term in RRF matter, and what would happen at `k=0`?

Sign in or go Pro to see the approach for this follow-up.

QHow does RRF interact with retrievers that return ranked lists of very different lengths?

Sign in or go Pro to see the approach for this follow-up.

QCan you derive a theoretical justification for `k=60`?

Sign in or go Pro to see the approach for this follow-up.

Don't say this

Red flags and common mistakes that signal junior thinking. Click to expand.

Most common mistake

Forgetting that RRF works on ranks, not raw scores. Replacing rank with the cosine or BM25 score gives a fundamentally different (and worse) fusion.

Sign in or go Pro to see all red flags and common mistakes. See Pro pricing.

60-second cheat sheet

The night-before-the-interview bullets. Scan these on the way to the call.

Sign in for 5 unlocks/day or go Pro for unlimited revision sheets. See Pro pricing.

References

Primary sources. Skim if you want the original framing.