Runs your real task examples against local quantized models via Ollama and recommends the smallest quant that holds quality.

Brutalist UX micro-library with quantized motion — 13 modules, zero dependencies. Magnetic buttons that snap to a 4px grid, mechanical tilt, dot-matrix text scramble, cursor trail on a 12px grid.

Run Gemma 3 locally in the browser via WebGPU. Q8_0 quantized, streaming generation, multi-turn chat with KV cache reuse.

Transform UUIDs into memorable 4-square visual patterns with quantized colors and emojis

The slides framework: a deck of viewport-sized slides moved one frame at a time as an ordinary aiui app — current slide + scene step as controls, wheel/touch quantized to one step per gesture, additive in-slide scenes, HUD grid overview with live previews

On-the-fly meshing of raster elevation tiles for the CesiumJS virtual globe

Fast on-device binary (sfw/nsfw) image classification for the browser — ~65 ms per image on CPU from a sub-1 MB-gzipped model. A uint8-quantized MobileNetV4-conv-small-050 runs through onnxruntime-web at 160px, tuned for pixel art, in a Web Worker with an

A nodejs cli version of https://github.com/rilden/tiledpalettequant (https://rilden.github.io/tiledpalettequant/) allowing for conversion of images to quantized bitmap files, with optional dithering

Emoji data and images

GIS data parser and converter CLI/library for Shapefile, MapInfo TAB, GeoJSON, KML, GPX, TopoJSON, CZML, CSV, ESRI JSON, MIF, GeoPackage, GML, FlatGeobuf, Cesium terrain tiles (terrain-RGB PNG, quantized-mesh, 3D Tiles b3dm) with multi-CRS support and str

Derived from mbostock's world-atlas: Pre-built TopoJSON from Natural Earth.

On-device pre-classification for Mushi Mushi reports using a quantized small language model running in onnxruntime-web (WASM/WebGPU). Filters obvious junk before reports leave the browser, cutting LLM cost and protecting user privacy.

Emoji data and images - apple

Emoji data and images - google

Emoji data and images - twitter

Cryptographic verification for quantized AI models. Proves a GGUF/AWQ/GPTQ matches the original full-precision model. ECDSA P-256.

🚀 **Techy & Futuristic** - QuantaKit delivers quantized UI elements in an Angular-based toolkit built for scale. With atomic design, blazing performance, and WCAG accessibility, it's your launchpad to consistent, adaptive apps—now and as Web Components i

Pre-built TopoJSON from the Ministry of the Interior.

> Part of the [VC Map Project](https://github.com/virtualcitySYSTEMS/map-ui)

No description provided.