npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@qhago/ai

v0.1.0-beta.1

Published

Optional AI barcode detector plugin for @qhago/scanner (ONNX Runtime Web, WebGPU/WASM).

Downloads

23

Readme

@qhago/ai

Beta experimental (npm i @qhago/ai@beta onnxruntime-web): aún no supera el gate G5 (falta medir la latencia en Android de gama media y validar con códigos 2D reales) — ver MODEL_CARD.md. En fotos públicas reales sube la tasa de lectura de 80,9 % a 87,0 % (1 lectura falsa en 818 fotos).

Detector de códigos opcional para @qhago/scanner. Cuando el motor clásico no lee nada durante unos frames, una red pequeña (CenterNet-lite, 0,22 M parámetros, 0,24 GMAC, 320×320) localiza los códigos y sus 4 esquinas; el motor rectifica cada región en su resolución original y la vuelve a leer.

import { createScanner } from "@qhago/scanner";
import { aiDetector, aiWorker } from "@qhago/ai";

const scanner = createScanner({
  workerFactory: aiWorker,        // worker con el motor + el detector (ONNX Runtime Web)
  plugins: [aiDetector()],        // "fallback" por defecto
});
await scanner.start();

También funciona con decode(image, { workerFactory: aiWorker, plugins: [aiDetector()] }).

Versiones compatibles

@qhago/ai declara @qhago/scanner como peer con el rango >=0.1.0-beta.0 <2.0.0 || >=1.0.0-0 <2.0.0. Admite las betas y RC actuales, 1.0.0-beta/1.0.0-rc y toda la 1.x. Con @qhago/[email protected], npm rechaza (ERESOLVE) @qhago/[email protected]: actualiza a 0.1.0-beta.1 o posterior.

Opciones

| Opción | Default | | |---|---|---| | when | "fallback" | "always" corre el detector en cada frame | | minConfidence | 0.4 | umbral del heatmap | | maxObjects | 5 | regiones por frame | | backend | "wasm" | "webgpu" / "auto" usan el modelo FP16 y el runtime WebGPU | | modelUrl | modelo del paquete | self-hosting / CDN | | wasmPaths | el de onnxruntime-web | carpeta de los .wasm de ORT | | intervalMs | 300 | mínimo entre inferencias en vídeo (la red cuesta ~40 ms en WASM, escritorio) | | cache | true | guarda el modelo en Cache API |

Peso

| Recurso | Tamaño | |---|---| | detector.int8.onnx (WASM) | ~0,33 MB | | detector.fp16.onnx (WebGPU) | ~0,45 MB | | runtime ORT WASM | ~3,7 MB gzip | | runtime ORT WebGPU | ~6,7 MB gzip |

Solo se descargan si usas el plugin. Sin @qhago/ai, @qhago/scanner no cambia de tamaño.

Bundlers

  • Vite: worker: { format: "es" } y optimizeDeps: { exclude: ["@qhago/scanner", "@qhago/ai", "onnxruntime-web"] } (ver apps/examples/ai-vite).
  • webpack 5: los new URL(…, import.meta.url) (worker, modelo, .wasm) se resuelven como assets sin configuración (ver apps/examples/ai-webpack; solo experiments.topLevelAwait por el ejemplo).

onnxruntime-web es peer dependency (^1.22): instálalo junto a este paquete.

Datos, métricas y limitaciones del modelo: MODEL_CARD.md.