Decoder for the Quantized Mesh 1.0 format

No description provided.

RaBitQ 1-bit quantized vector index in WebAssembly — 32× embedding compression with high-recall rerank, for browsers, Cloudflare Workers, Deno, and Bun

Flexible quantized transation framework

WebLLM provider for @localmode - LLM inference with quantized models

Decoder for the Quantized Mesh 1.0 format

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

Fixed‑timestep update loop with independent draw rate and quantized interpolation

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.

Tiny offline toxicity classifier for the browser — correctly int8-quantized bert-tiny-toxicity (~4.5 MB) over onnxruntime-web. Nothing ever leaves the device.

Lightweight French lemmatization and conjugation. Convert between nouns, verbs, and adjectives using fast, quantized INT8 ONNX models under 2MB.

JS SDK for running quantized small language models in the browser

Vector search for JSON datasets. Build quantized indexes and search with WASM SIMD.

Pre-built TopoJSON from the U.S. Census Bureau.

A [pi](https://github.com/badlogic/pi) extension that overrides the built-in `edit` tool with whitespace-tolerant matching, designed for local/quantized LLMs.

A Node.js library that converts DEM meshes (DXF, GeoJSON, etc.) into Cesium Terrain Tiles (quantized-mesh or heightmap) with flexible zoom-level control and multi-threading support.

Datasource to render terrain data in quantized-mesh and terrarium formats with harp.gl

RaBitQ 1-bit quantized vector index in WebAssembly — 32× embedding compression with high-recall rerank, for browsers, Cloudflare Workers, Deno, and Bun

Internal component of Sutra SDK — applications should install sutra-sdk. The Edge Harness. Offline-first autonomous agent runtime for local SLMs (Phi-3, Gemma, quantized GGUF/ONNX) with durable friction telemetry and on-device vector memory.

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