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

@ruvector/tiny-dancer-wasm

v0.1.0

Published

WebAssembly bindings for Tiny Dancer - FastGRNN neural inference for AI routing in browsers

Downloads

71

Readme

Ruvector Tiny Dancer WASM

npm Crates.io License: MIT

WebAssembly bindings for Tiny Dancer neural routing.

ruvector-tiny-dancer-wasm brings production-grade AI agent routing to the browser with WebAssembly. Run FastGRNN neural inference for intelligent request routing directly in client-side applications. Part of the Ruvector ecosystem.

Why Tiny Dancer WASM?

  • Browser Native: Run neural routing in any browser
  • Low Latency: Sub-millisecond inference times
  • Small Bundle: Optimized WASM binary (~100KB gzipped)
  • Offline Capable: No server required for inference
  • Privacy First: Route decisions stay client-side

Features

Core Capabilities

  • Neural Inference: FastGRNN model execution
  • Feature Engineering: Request feature extraction
  • Multi-Agent Routing: Score and rank agent candidates
  • Model Loading: Load pre-trained models
  • Batch Inference: Process multiple requests

Advanced Features

  • Web Workers: Background inference threads
  • Streaming: Process streaming requests
  • Model Caching: IndexedDB model persistence
  • Quantization: INT8 models for smaller size
  • SIMD: Hardware acceleration when available

Installation

npm install @ruvector/tiny-dancer-wasm
# or
yarn add @ruvector/tiny-dancer-wasm

Quick Start

Basic Usage

import init, { TinyDancer, RouteRequest } from '@ruvector/tiny-dancer-wasm';

// Initialize WASM module
await init();

// Create router instance
const router = new TinyDancer();

// Load pre-trained model
await router.loadModel('/models/router-v1.bin');

// Create routing request
const request: RouteRequest = {
  query: "What is the weather like today?",
  context: {
    userId: "user-123",
    sessionLength: 5,
    previousAgent: "general",
  },
  agents: ["weather", "general", "calendar", "search"],
};

// Get routing decision
const result = await router.route(request);
console.log(`Route to: ${result.agent} (confidence: ${result.confidence})`);

With Web Workers

import { TinyDancerWorker } from '@ruvector/tiny-dancer-wasm/worker';

// Create worker-based router (non-blocking)
const router = new TinyDancerWorker();

// Initialize in background
await router.init();
await router.loadModel('/models/router-v1.bin');

// Route without blocking main thread
const result = await router.route(request);

Feature Engineering

import { FeatureExtractor } from '@ruvector/tiny-dancer-wasm';

const extractor = new FeatureExtractor();

// Extract features from request
const features = extractor.extract({
  query: "Book a flight to Paris",
  tokens: 6,
  language: "en",
  sentiment: 0.7,
  entities: ["Paris"],
});

console.log(`Feature vector: ${features.length} dimensions`);

API Reference

TinyDancer Class

class TinyDancer {
  constructor();

  // Model management
  loadModel(url: string): Promise<void>;
  loadModelFromBuffer(buffer: Uint8Array): void;

  // Routing
  route(request: RouteRequest): Promise<RouteResult>;
  routeBatch(requests: RouteRequest[]): Promise<RouteResult[]>;

  // Scoring
  scoreAgents(request: RouteRequest): Promise<AgentScore[]>;

  // Info
  getModelInfo(): ModelInfo;
  isReady(): boolean;
}

Types

interface RouteRequest {
  query: string;
  context?: Record<string, any>;
  agents: string[];
  constraints?: RouteConstraints;
}

interface RouteResult {
  agent: string;
  confidence: number;
  scores: Record<string, number>;
  latencyMs: number;
}

interface AgentScore {
  agent: string;
  score: number;
  features: number[];
}

interface RouteConstraints {
  excludeAgents?: string[];
  minConfidence?: number;
  timeout?: number;
}

Bundle Optimization

Tree Shaking

// Import only what you need
import { TinyDancer } from '@ruvector/tiny-dancer-wasm/core';
import { FeatureExtractor } from '@ruvector/tiny-dancer-wasm/features';

CDN Usage

<script type="module">
  import init, { TinyDancer } from 'https://unpkg.com/@ruvector/tiny-dancer-wasm';

  await init();
  const router = new TinyDancer();
</script>

Performance

Benchmarks (Chrome 120, M1 Mac)

Operation           Latency (p50)
────────────────────────────────
Model load          ~50ms
Single inference    ~0.5ms
Batch (10)          ~2ms
Feature extraction  ~0.1ms

Bundle Size

Format              Size
────────────────────────
WASM binary         ~100KB gzipped
JS glue             ~5KB gzipped
Total               ~105KB gzipped

Browser Support

| Browser | Version | SIMD | |---------|---------|------| | Chrome | 89+ | ✅ | | Firefox | 89+ | ✅ | | Safari | 15+ | ✅ | | Edge | 89+ | ✅ |

Related Packages

Documentation

License

MIT License - see LICENSE for details.


Part of Ruvector - Built by rUv

Star on GitHub

Documentation | npm | GitHub