winkjs/wink-nlp
Developer friendly Natural Language Processing ✨
Provides comprehensive NLP pipeline (tokenization, POS tagging, NER, sentiment analysis, entity recognition) with integrated word embeddings for semantic tasks like similarity and classification. Built as zero-dependency (~10KB gzipped) with pre-trained compact language models (~1MB) and runs across Node.js, browsers, and Deno with full TypeScript support. Processes text at 650K+ tokens/second while supporting multilingual tokenization and offering BM25 vectorization, cosine similarity, and text preprocessing utilities.
1,356 stars. Used by 5 other packages. Available on npm.
Stars
1,356
Forks
64
Language
JavaScript
License
MIT
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Reverse dependents
5
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/winkjs/wink-nlp"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Compare
Related tools
LSYS/LexicalRichness
:smile_cat: :speech_balloon: A module to compute textual lexical richness (aka lexical diversity).
mbejda/Node-OpenNLP
Apache OpenNLP wrapper for Nodejs
LanguageMachines/frog
Frog is an integration of memory-based natural language processing (NLP) modules developed for...
winkjs/wink-lexicon
English lexicon useful in NLP/NLU
winkjs/wink-nlp-utils
NLP Functions for amplifying negations, managing elisions, creating ngrams, stems, phonetic...