Show HN: TurboQuant-WASM – Google's vector quantization in the browser

TurboQuant WASM

Experimental WASM + relaxed SIMD build of botirk38/turboquant for browsers and Node.js.

Based on the paper "TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate" (Google Research, ICLR 2026).

Live Demo — vector search, image similarity, and 3D Gaussian Splatting compression running in the browser.

What this adds

npm package with embedded WASM — npm install turboquant-wasm

with embedded WASM — Relaxed SIMD — @mulAdd FMA maps to f32x4.relaxed_madd

— FMA maps to SIMD-vectorized QJL sign packing/unpacking and scaling

QJL sign packing/unpacking and scaling TypeScript API — TurboQuant.init() / encode() / decode() / dot()

— / / / Golden-value tests — byte-identical output with the reference Zig implementation