Vecnest

Vecnest is an open-source, browser-based vector database. Vectors are stored in IndexedDB and queried via cosine similarity. Optional Transformers.js embeddings let you add and search by text—no backend required.

---

Features

Feature	Description
IndexedDB persistence	Vectors and metadata survive reloads; same-origin storage.
HNSW indexing	Approximate nearest neighbor search via WebAssembly for O(log n) performance.
Cosine similarity search	Fast vector search with automatic HNSW optimization.
Optional embeddings	Use addText / searchText with Transformers.js, or supply your own vectors.
Update & delete	Update vectors and metadata by ID; delete by ID with optional HNSW rebuild.
Metadata filtering	Filter search results with operators: $eq, $gt, $gte, $lt, $lte, $in, $nin.
Web Workers	Background-thread search to keep the UI responsive.
React hooks	useVecnest, useSearch, useSearchText for React 18+.
List & count	List vectors with pagination; count total; storage estimate.

---

Quickest path

1. Install: npm install vecnest
2. Connect: await db.connect()
3. Add & search: db.addText(...) and db.searchText(...)
4. Done. No backend.

import { VecnestDB } from 'vecnest';

const db = new VecnestDB('my-db');
await db.connect();

await db.addText('Machine learning is a subset of AI.', { source: 'doc1' });
await db.addText('Neural networks learn from data.', { source: 'doc2' });

const results = await db.searchText('What is ML?', { k: 5 });
console.log(results); // [{ id, score, metadata }, ...]

db.close();

---

Documentation

• Installation — Requirements, install, and peer dependencies.
• Quick start — Minimal JavaScript and React examples.
• Vecnest UI setup — Add the inspector (DBs, docs, 3D) to your app.
• Guide — Step-by-step: JavaScript API, React hooks, CRUD, search.
• API reference — Full API with code snippets.
• Metadata filtering — Filter syntax and examples.
• Architecture — Storage, search, embeddings, SDK design.
• Examples — Demo, JS-only, and RAG vanilla/React examples.

---

Use cases

• Offline RAG — Semantic search over your data without a server.
• Local-first knowledge bases — Store and query documents entirely in the browser.
• Privacy-preserving chat — Embed and search over user data on-device.
• Prototypes & demos — Add vector search to a frontend without deploying a backend.

---

License

AGPL-3.0. See LICENSE in the repository.