v148 · AI · JavaScript

Quality Control Panel

Benchmark the Translator API across multiple language pairs simultaneously. See per-pair latency, availability status, and translation output — all from a single source text. Identify which pairs are downloaded and ready vs. need a model download.

—Avg latency (ms)

—Pairs available

—Inference mode

Source text & pairs

Source language

Source text

Translation pairs

// Check availability before creating — avoids silent failures
const PAIRS = [
  { from: 'en', to: 'fr' },
  { from: 'en', to: 'de' },
  { from: 'en', to: 'ja' },
];

const results = await Promise.all(PAIRS.map(async ({ from, to }) => {
  const avail = await Translator.availability({ sourceLanguage: from, targetLanguage: to });
  if (avail === 'no') return { pair: `${from}→${to}`, error: 'unavailable' };

  const t = await Translator.create({ sourceLanguage: from, targetLanguage: to });
  const t0 = performance.now();
  const text = await t.translate(sourceText);
  return { pair: `${from}→${to}`, text, latency: performance.now() - t0 };
}));

implementation reference

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

see also

implementation reference