Smallest Pulse Author Benchmark - East Asian streaming - Mandarin - CV-25: leaderboard
Smallest Pulse Author Benchmark - East Asian streaming - Mandarin - CV-25 — Smallest AI Pulse own author model-card evaluation, content hash pinned. Native model=pulse; never Pulse Pro or Pulse2. Comparator observations with unbound product versions retained as evaluation runs, not public model releases. Native percentage values unchanged; error rates unbounded above100. Source explicitly labels WER, not CER. US-region WebSocket endpoint wss://api.us.smallest.ai/waves/v1/stt/live?model=pulse; native CV25/FLEURS/language-specific corpus. Preserve language and dataset; average is native12dataset mean. Metric: WER (%). Source: https://docs.smallest.ai/models/model-cards/speech-to-text/pulse
Metric: WER (%). Source: docs.smallest.ai. Saturation forecast: Rough model projection: around 2026. 2 models tracked. Updated 16 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Smallest AI Pulse | 15.99 | |
| 2 | Deepgram Nova 3 | 22.44 |
Interactive version: theaggregate.ai/benchmark?slug=smallest-pulse-author-benchmark-east-asian-streaming-mandarin-cv-25 · How It Works · Data refreshed daily, snapshot 2026-11-16.