BM25
The keyword-ranking algorithm behind classic search engines.
/ quick answer
BM25 scores documents by term frequency, inverse document frequency, and length normalization. It's fast, deterministic, and complements dense vector retrieval in hybrid search. The keyword-ranking algorithm behind classic search engines.
/ frequently asked
What is BM25?
BM25 scores documents by term frequency, inverse document frequency, and length normalization. It's fast, deterministic, and complements dense vector retrieval in hybrid search.
What is an example of BM25?
Elasticsearch and Postgres full-text search both use BM25 under the hood.
Why does BM25 matter?
The keyword-ranking algorithm behind classic search engines.
What should I learn after BM25?
Open the related workflows and prompts further down this page: they show how this idea is applied in a real task, step by step.
/ continue exploring
Related concepts
The vocabulary this page depends on.
- →Semantic Search
Finding information by meaning rather than exact keyword match.
- →Retrieval
Selecting the most relevant chunks for a query before generation.
- →Hybrid Search
Combining keyword (BM25) and vector search for better recall.
- →RAG (Retrieval-Augmented Generation)
Inject external knowledge into an LLM at query time.
Related workflows
Turn this into a repeatable process.
- →AI Search Ranking Audit
An AI Search Ranking Audit evaluates digital content and technical infrastructure to identify strengths and weaknesses in its ability to be accurately discovered, understood, and ranked by generative AI systems.
- →Automated Competitor Research
From a product description to a structured competitor matrix in under 10 minutes.
- →Build an Internal Knowledge Bot
Ship a Slack bot that answers questions from your company docs.
- →AI Content Factory: One Topic to Ten Assets
Convert a single topic into a full multi-channel content drop.
Related prompts
Reusable prompts for this job.
- →Competitor Discovery Prompt
Surface and structure direct competitors for a given product.
- →Grounded Answer Prompt
Force the model to answer only from provided sources, with citations.
- →Viral Hook Generator Prompt
Produce 10 scroll-stopping hooks for a topic and platform.
Comparisons & alternatives
Pick between the options.
- →Perplexity vs ChatGPT Search
Two answer engines, two philosophies: citation-first research vs conversational assistant.
- →Perplexity vs You.com vs SearchGPT
Three AI-first search engines compared.
- →RAG vs Fine-Tuning
When to retrieve, when to retrain.
- →Zapier vs Make (Integromat)
Which no-code automation platform fits your operation.