WASM-NATIVE VECTOR DATABASE

EDGEVEC

High-performance vector similarity search that runs entirely in the browser. HNSW indexing, binary quantization, and SIMD acceleration.

32x
Memory Reduction
<1ms
Search Latency
2x+
SIMD Speedup
EXPLORE DEMOS VIEW ON GITHUB

BINARY QUANTIZATION

32x memory reduction with minimal recall loss. Store millions of vectors in browser memory.

SIMD ACCELERATION

WASM SIMD128 optimized distance calculations. 2x+ faster Hamming distance with @jsonMartin's contribution.

METADATA FILTERING

SQL-like query syntax for pre/post filtering. Combine vector search with structured queries.

PURE WASM

No server required. Runs entirely in browser or Node.js. Zero external dependencies.

HNSW INDEXING

Hierarchical Navigable Small World graphs for logarithmic search complexity.

PERSISTENCE

Save and load indexes. IndexedDB support for browser storage. Snapshot/restore capability.