embedding-utils
Vector math, similarity search, ANN indexing, clustering, async pipelines, evaluation metrics, and multi-provider embedding generation -- zero dependencies, full TypeScript, one import.
A zero-dependency, full-TypeScript library that bundles vector math, similarity search, ANN indexing, clustering, async pipelines, evaluation metrics, and multi-provider embedding generation behind a single import. It is built to support semantic search, RAG pipelines, recommendation engines, duplicate detection, and document clustering without pulling in heavy ML frameworks or vector databases. Capabilities include HNSW approximate nearest-neighbor search, hybrid search via reciprocal rank fusion (RRF), HDBSCAN clustering, quantization, dimensionality reduction through random projection, and markdown-aware chunking. It has zero production dependencies.
Installation
Requires Node.js 18+. Supports both ESM and CommonJS.
npm install embedding-utilsFor local ONNX inference (no API key needed), also install the optional peer dependency:
npm install @huggingface/transformersRelated \\
-
chunk-match
A NodeJS library that semantically chunks text and matches it against a user query using cosine similarity for precise and relevant text retrieval.
-
extract-topics
Use LDA (Latent Dirichlet Allocation) to extract topics from text
-
fast-topic-analysis
A tool for analyzing text against predefined topics using average weight embeddings and cosine similarity.