MaartenGr/KeyBERT

Health v2 · maintenance only

Adoption not part of the score

4215 stars · 386 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

KeyBERT is a minimal Python library that extracts keywords and keyphrases from documents using BERT embeddings and cosine similarity. It requires no training and can be used with just a few lines of code, with optional MMR and Max Sum Distance for diversity.

Use cases

• extract keywords from a document
• find keyphrases that best describe a text
• generate tags for articles automatically
• get diverse keywords using MMR
• use custom transformer models for keyword extraction
• index documents by their most representative phrases

When to choose

• you need quick, training-free keyword extraction with transformer embeddings
• you want a simple pip-installable Python API with minimal code
• you need diverse keyphrases via MMR or Max Sum Distance
• you want to plug in different embedding backends like Sentence-Transformers, Flair, spaCy, or Gensim

When to avoid

• you need supervised, domain-specific keyphrase extraction with high precision
• you cannot run transformer models due to compute constraints
• you need a full NLP pipeline rather than just keyword extraction

Facets

library · maturity stable

nlp machine-learning search-engine machine-learning python keyword-extraction keyphrase-extraction bert embeddings mmr cosine-similarity natural-language-processing search

Member repositories

For agents

markdown · JSON · MCP: product_card(name="MaartenGr/KeyBERT")