v148 · AI · JavaScript

Quality Control Panel

Benchmark the Language Detector API against a heuristic fallback on 10 diverse text samples. See detection latency, confidence scores, and whether the on-device model is available — all in one panel.

— Avg latency (ms)

— Correct detections

— Inference mode

Benchmark

checking…

// Benchmarking the Language Detector API
async function benchmark(texts) {
  const detector = await LanguageDetector.create({
    expectedInputLanguages: texts.map(t => t.expected)
  });
  const results = [];

  for (const { text, expected } of texts) {
    const t0 = performance.now();
    const [top] = await detector.detect(text);
    const latency = performance.now() - t0;
    results.push({
      detected: top.detectedLanguage,
      confidence: top.confidence,
      correct: top.detectedLanguage === expected,
      latencyMs: latency.toFixed(1),
    });
  }
  return results;
}

references

implementation reference

Need the exact API surface, compatibility boundaries, errors, lifecycle, and source links? Read the matching gendn reference ↗

see also

references

implementation reference